Académique Documents
Professionnel Documents
Culture Documents
Competitive Advantage
through Collaborative
Innovation Networks
Peter A. Gloor
OXFORD UNIVERSITY PRESS
SWARMCREATIVITY
This page intentionally left blank
SwarmCreativity
Competitive Advantage through
Collaborative Innovation Networks
Peter A. Gloor
2006
Oxford University Press, Inc., publishes works that further
Oxford Universitys objective of excellence
in research, scholarship, and education.
Oxford New York
Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With ofces in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam
Copyright 2006 by Peter A. Gloor
Published by Oxford University Press, Inc.
198 Madison Avenue, New York, New York 10016
www.oup.com
Oxford is a registered trademark of Oxford University Press
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise,
without the prior permission of Oxford University Press.
Library of Congress Cataloging-in-Publication Data
Gloor, Peter A. (Peter Andreas), 1961
Swarm creativity : competitive advantage through collaborative
innovation networks / Peter A. Gloor.
p. cm.
Includes bibliographical references and index.
ISBN-13: 978-0-19-530412-1
ISBN 0-19-530412-8
1. Business networks. 2. Information networks. 3. Group decision making.
4. Teams in the workplace. 5. Creative ability in business. 6. Technological
innovationsManagement. 7. Knowledge management. I. Title.
HD69.S8.G586 3006
338.8'7dc22 2005011977
9 8 7 6 5 4 3 2 1
Printed in the United States of America
on acid-free paper
ACKNOWLEDGMENTS
It is tting that a book about swarm creativity and collaborative innovation
networks (COINs) should result from a COIN that itself worked in swarm
creativity. This book is not the work of a single person, but resulted from the
interactions of a large and wonderful group of people. Without the generous
help of this group of friends, it would never have been completed.
Several individuals deserve mention. My rst thanks go to Thomas Ma-
lone, my generous host for the last three years at the Massachusetts Institute
of Technology (MIT) Center for Coordination Science (CCS), and to Thomas
Allenof the MITSloanSchool of Management. Bothhave beenmost inspiring
mentors. I am also grateful to Rob Laubacher, John Quimby, George Her-
mann, and Peggy Nagel, all colleagues at CCS, for many thought-provoking
discussions.
Yan Zhao developed countless versions of the temporal communication
ow analysis (TeCFlow) software tool, putting up with myriad requests for
change and always adding her own excellent ideas and insights. Our work at
Dartmouth College has been generously supported by Hans Brechbuhl and
M. Eric Johnson at the Tuck School of Business Center for Digital Strategies,
and by Fillia Makedon from the Dartmouth Devlab.
Myformer colleagues at Deloitte have beeninstrumental indevelopingthe
original concept of collaborative knowledge networks (CKNs) and COINs. I
am deeply indebted to Robin Athey, cofounder of the rst COIN on CKNs.
Elmar Artho, Jessica Bier, Emma Connolly, Niki Flandorfer, Anne Gauton,
Laura Ghezzi, Karin Guentensperger, Adriaan Jooste, Roland Haenni, Marc
acknowledgments
Killian, Nico Kleyn, Steven Lambert, Thomas Ojanga, Thomas Ritzi, Stuart
Rosenberg, Martin Spedding, Andreas Timpert, Gilbert Toppin, Virginia
Villalba, andmost of allThomas Schmalberger were crucial contributors
to this COIN. I am also truly grateful to my colleagues in the earliest COINs.
Brian Dunkel, Scott Dynes, Irene Lee, and Angel Velez-Sosa in the Animated
Algorithms project, Rade Adamov, Brian Dunkel, Norbert Hoffmann, Peter
Huegli, Wolfgang Luef, Zoltan Majdik, Rudolf Marty, and Kurt Wolf at
UBS, and Stephan Carsten, Emanuele Fossati, Sandra Giacalone, Hans Peter
Hochradl, Marc Killian, Till Knorr, Rudolf Lehmann, Franz Odekerken,
Christoph Sand, Roland Stadler, Ragnar Wachter, and Jill Williams at Deloitte
were inuential in developing my early understanding of the workings of
COINs.
My friends and colleagues Hermann Blaser, Scott Dynes, Bill Ives, Takis
Metaxas, Andre Ruedi, Suzanne Spencer-Wood, and Wayne Yuhasz helped
me through the ups and downs of the gestation process, reading and com-
menting on various versions of the manuscript, offering great advice, and
contributing their own excellent ideas.
I amgrateful to the late David Hoenigsberg, Douglas Grannell, and Gareth
Dylan Smith; to Fritz Bircher and Reinhold Krause; to Christoph Von Arb and
Pascal Marmier; and to Richard E. Deutsch, Martin Duerst, Erich Gamma,
Stefano Mazzocchi, and David Tennenhouse for sharing the insights they
gained in leading their own COINs.
Cathy Benko, Rob Cross, Doug Downing, Walter Etter, Jan Fuelscher,
Dirk Havighorst, Carey Heckmann, Bruce Hoppe, Marcel Isenschmid, Char-
les Leiserson, Bernardo Lindemann, Stefano Pelagatti, Gregor Schrott, Gary
Shepherd, Marshall Van Alstyne, and Daniel Wild all offered excellent input
in their reviews of earlier versions of this manuscript.
I am deeply indebted to Scott Cooper for adding his own insights while
making my writing much more accessible. I am also grateful to John Rausch-
enberg at Oxford University Press for his support in publishing this book.
This book is dedicated to my children, Sarah and David, my best teachers
about what matters.
CONTENTS
Introduction: At the Tipping Point 3
1 COINs and Their Benets 9
2 Collaborative Innovation through Swarm Creativity 19
3 The DNA of COINS: Creativity, Collaboration,
and Communication 49
4 Ethical Codes in Small Worlds 71
5 Real-Life Examples: Lessons Learned from COINs 91
6 COINs and Communications Technology 115
appendixes 125
A Collaborative Knowledge Networks (CKNs) 127
B Temporal Communication Flow Analysis (TeCFlow) 141
C Knowledge Flow Optimization (KFO) 173
Notes 193
References 199
Additional Resources 203
Index 209
This page intentionally left blank
SWARMCREATIVITY
This page intentionally left blank
INTRODUCTION
At the Tipping Point
As managers, we need to shift our thinking from command
and control to coordinate and cultivatethe best way to gain
power is sometimes to give it away.
Thomas W. Malone, The Future of Work (2004)
We are at the dawn of a new way of working together, thanks in large part to
technological advances that allow a more radically superior mode of innova-
tion than ever before. Soon, businesses throughout the world will be looking
for ways to unleash the power of their COINs. Some will be scrambling to
gure out how to innovate in the new business environment.
Few today know COINscollaborative innovation networksby that
name, even though COINs have been around for hundreds of years. Many of
us have already collaborated in COINs without even knowing it.
What makes them so relevant today is they have reached their tipping
point.
This idea of a tipping point comes fromMalcolmGladwell, who describes
the moment of critical mass where radical change is more than just possible,
but is a certainty. Thanks to the communication capabilities of the Internet,
COINs are at that threshold; they have achieved global reach and they can
spread at viral speed. Gladwell uses the word epidemic to describe what
happens at the tipping point.
1
3
swarm creativity
4
What is a COIN? There is a lot to the answer, and this book delves deeply
into the DNA of COINsthe traits that must exist for them to exist and for
them to succeed. Here, we offer a brief denition:
A COIN is a cyberteam of self-motivated people with a collective
vision, enabled by the Web to collaborate in achieving a common goal
by sharing ideas, information, and work.
In a COIN, knowledge workers collaborate and share in internal transparency.
They communicate directly rather than through hierarchies. And they inno-
vate and work toward common goals in self-organization instead of being
ordered to do so. Working this way is key to successful innovation, and it is
no exaggeration to state that COINs are the most productive engines of inno-
vation ever. COINs have produced some of the most revolutionary drivers of
change of the Internet age, such as the World Wide Web and Linux.
This book explains COINs in depth, makes the case for why businesses
ought to be rushing to uncover their COINs and nurture them, and pro-
vides tools for building organizations that are more creative, productive, and
efcient by applying principles of creative collaboration, knowledge shar-
ing, and social networking. It contains information on leveraging COINs to
develop successful products in research and development, grow better cus-
tomer relationships, establish better project management processes, and build
higher-performing teams. It even offers a method for locating, analyzing, and
measuring the impact of COINs on an organization.
In short, this book answers four key questions:
1. Why are COINs better at innovation than conventional organizations?
2. What are the key elements of COINs?
3. Who are the people that participate in COINs, and how do they become
COIN members?
4. How does an organization transform itself into a COIN?
In subsequent chapters, you will learn about the creators who come up with
the visionary ideas, the communicators who serve as ambassadors of COINs
and help carry newinventions over their tipping points, and the collaborators
who form the glue of a COIN and make it see the vision through to reality.
At the Tipping Point
5
In a sense, this is really two books in one. Chapters 1 through 6 make the
business case for COINs and explain the technological advances that have led
to their tipping point. Appendixes A through C are tools and steps to COINs:
how to uncover COINs by analyzing the evolution of communication pat-
terns; how to leverage COINs and combine customers and suppliers into a
seamlessly integrated value network; and how to reassign tasks of COINs.
One of these tools also shows individuals how to become efcient members
of COINs and fully leverage their own skills as creators, communicators, and
collaborators.
The book is backed by numerous case studies of different types. Principles
are illustrated by famous COINs, such as the groups that created the Web and
Linux, as well as by stories of how ordinary people are leveraging COINs
to achieve extraordinary success. The high-prole successes illustrate the
tremendous potential of COINs, and the everyday COINs demonstrate that
anyone caninitiate andsucceedwitha COIN. Chapter 5presents some specic
examples from a leading manufacturer, a global service provider, nancial
institutions, and other sectors.
InnovateCollaborateCommunicate
The creation of the World Wide Web is such an exemplary COIN that it
warrants some further mention in this introduction. When Tim Berners-Lee
introducedthe WorldWide Webtothe academic worldat the 1991Association
for Computing Machinery (ACM) Hypertext Conference in San Antonio,
Texas, the hypertext concept had already come a long way.
2
Aseries of visionaries before Berners-Lee proposed the basic ideas behind
the Web, namely to link pieces of information and make them accessible to
many users. The rst was Vannevar Bush, the famous scientist and advisor to
Franklin Delano Roosevelt. In a 1945 magazine article, he described a system
called Memex to make and followlinks.
3
But the information technology that
emerged in the 1950s consisted of microches and card readers, and Bushs
Memex system was not built in his lifetime. In the 1960s, other visionaries
moved the idea forward. Ted Nelson, who coined the term hypertext, and
computer scientist Douglas Engelbart, who demonstrated the rst working
hypertext system, took up Bushs idea. Engelbart invented the computer
swarm creativity
6
mouse while building the prototype oNLine System (NLS) that provided
hypertext browsing, editing, and e-mail. However, both men were still ahead
of their time, and the spark did not fully ignite.
In the late 1980s, the hypertext concept found a rm footing in the
academic computer science community, bringing together hundreds of re-
searchers at the annual hypertext conferences. But it was only in the early
1990s that technology and society were ready for hypertext, and a team of
enthusiasts collaborated to spin the Web.
Tim Berners-Lee wrote his rst hypertext systemEnquirein 1980,
while working as a consultant for CERN, the European Organization for
Nuclear Research laboratory. It took him years of grassroots lobbying at
CERN, writing and circulating research proposals, until his boss nally
approved the purchase of a NeXT computer in 1990s and allowed Berners-
Lee to go ahead and write a global hypertext system. During the same year,
Robert Cailliau, another hypertext enthusiast fromthe CERNcomputer group,
joined the effort. Together, they developed the rst Web browser, editor, and
server, soon reinforced by a small teamof volunteers, mostly summer students
at CERN.
Finally, in December 1991 at the San Antonio conference, Berners-Lee
and Cailliau presented the ideas of their group to the academic community
at large. By that time, the fervor had already spread, and the rst Web server
outside Europe had been installed. People ocked to the Web development
Figure I.1. From the Memex to the ubiquitous Internet. Memex photo from Bush,
As We May Think (1945); NeXt cube photo courtesy of NeXt, Inc.; Sony Ericsson
P900 photo courtesy of Sony Ericsson.
At the Tipping Point
7
team, and programmers fromFinland, Austria, Germany, France, and Califor-
nia built new versions of Web browsers and servers. In February 1993, when
Marc Andreessen of the National Center for Supercomputing Applications
(NCSA) at the University of Illinois released a browser called Mosaic, the
tipping point was reached. The Web exploded. The following year, Andreesen
and his colleagues left NCSA to form Mosaic Communications Corporation,
which later became Netscape and turned the twenty-somethings into very
young millionaires.
The birth and explosive growth of the Web exhibit all the characteristics
of a highly successful COIN at work. In the next chapters, well explore the
swarm creativity that resulted in the Web, learn how internal transparency
creates efciency and leads to success, and examine how virtual trust and an
ethical code hold a COIN together so it can fulll its vision.
To make the business case, though, we begin with benets in chapter 1.
COINs offer tremendous innovative power, and well see that Thomas Mal-
ones words from the beginning of this introduction are true. If working col-
laboratively, in a transparent environment, is giving away power, it is also
the way to gain the power of COINs.
This page intentionally left blank
CHAPTER1
COINs and Their Benefits
Late in 1999, I was riding the train in Switzerland from Basel to Zurich when
my mobile phone rang. A colleague of mine wanted to know if I could be on
a conference call, right away, with some people from DaimlerChrysler. They
had a problem.
Gary Valade, DaimlerChryslers executive vice president and head of its
Global Procurement and Supply division, was facing a big challenge. He
wanted to capitalize on emerging technologies and streamline his function,
optimize transaction processes, tighten control of materials ow and inven-
tory, andimprove the transfer of informationthroughout the companys supply
chain. He needed to do this in order to address the competitive pressure from
other players in the automotive industry; DaimlerChryslers competitors were
also busy leveraging e-business technologies in automotive procurement and
supply chain processes. Whatever solution was developed had to be imple-
mented on a worldwide basis.
DaimlerChrysler ultimately solved the problem and answered the chal-
lenge with its e-extended Enterprise (e
3
) initiative. This initiative has en-
hanced DaimlerChryslers supplier relationships, increased efciencies,
reduced new product development time, and cut expenses. Well learn more
about the project in chapter 5. What is important here is how the project
unfolded.
9
swarm creativity
10
The e
3
project began in January 2000. At DaimlerChrysler, some 15
people from the Global Procurement and Supply function worked full-time,
supported by members of the information technology (IT) organization and
by part-time help from employees of the business units, as required. A team
of 15 external consultants from the United States and Europe reinforced the
project team. This effort led to a successful outcome: the company spent
slightly more than $10 million to optimize its procurement function, and it
saved more than $50 million in the rst year of operation.
Although it wasnt planned from the beginning as such, the project team
operated as a global collaborative innovation network, or COIN.
DaimlerChrysler met a tremendous challenge successfully, thanks in large
part to a new way of working together in a business environment driven
more by knowledge than ever before. Today, collaborating in an open and
transparent ow of knowledge is key to successful innovation, and COINs
are the most productive engines of innovation ever. COINs have produced
some of the most revolutionary drivers of change in the Internet age. One
example is Linux. Another is the World Wide Web itself.
What makes COINs better? They allow for building organizations that
are more creative, productive, and efcient by applying principles of cre-
ative collaboration, knowledge sharing, and social networking. Sponsors and
members of COINs often change their work and leadership styles to become
more creative innovators, more efcient communicators, and more produc-
tive collaborators. The evidence shows that COINs can be leveraged to de-
velop successful products in research and development (R&D), grow better
customer relationships, establish better project management processes, and
build high-performing teams. COIN-enabled organizations demonstrate more
efcient leadership, culture, structure, and business processes.
How did it happen that the team from Switzerlanda small landlocked
countrywon the 2003 Americas Cup, sailings most coveted prize? The
answer illustrates the three words that make up the COIN acronym. Alinghi,
the Swiss sailing team, functioned as a close-knit, collaborative community
of more than 100 incredibly motivated experts and athletes from15 countries.
Ernesto Bertarelli, the Swiss biotechnology billionaire who provided the
nancial backing for the team, was just another team member. Alinghi
developed superior technology by creatively applying research insights from
other areas; consider, for instance, the teams innovation in uid dynamics,
COINs and Their Benefits
11
which came through collaboration with a Swiss university medical researcher
who was modeling the ow of blood vessels in the human body. Today,
Alinghi continues to generate and test its innovative ideas through a research
network of ve universities. Dirk Kramers, Alinghis chief designer, is very
clear in his praisenot for geniuses working in isolation, but for collaborative
networks of highly motivated researchers. Team Alinghi illustrates how
organizations operating as COINs become leaders of innovation, exibility,
and efciency.
COIN applications range from IT outsourcing, sales force optimization,
R&D, mergers and acquisitions, software, and distributed product develop-
ment to running political campaigns and charities online. Projects in compa-
nies such as DaimlerChrysler, Union Bank of Switzerland (UBS), Novartis,
Intel, Deloitte, HP, and IBM have successfully operated as COINs to deliver
innovations more rapidly and at lower costs. Youll learn about some of those
projects in later chapters.
What Makes a COIN?
Members of COINs self-organize as cyberteams, teams that connect people
through the Internetenabling themto work together more easily by commu-
nicating not through hierarchies, but directly with each other. The individuals
in COINs are highly motivated, working together toward a common goal
not because of orders from their superiors (although they may be brought
together in that way), but because they share the same goal and are convinced
of their common cause.
People in COINs usually assemble around a newidea outside of organiza-
tional boundaries and across conventional hierarchies. They innovate together
in swarmcreativitywhich youll learn about in the next chapter. By swarm-
ing together, they work together in a structure that enables a uid creation
and exchange of ideas. Looked at from the outside, the structure of a COIN
may appear chaotic, like a bee or ant colony, but it is immensely productive
because each team member knows intuitively what he or she needs to do.
COIN members develop new ideas as a team; the whole is much more
than the sum of its parts. They create and share knowledge in an environment
of high trust by collaborating under a common code of ethics.
swarm creativity
12
In short, three characteristics dene COINs. They:
1. Innovate through massive collaborative creativity
2. Collaborate under a strict ethical code
3. Communicate in direct-contact networks
COINs Deliver Benets to Organizations
The benets to organizations that embrace the concept of COINs are sub-
stantial.
COINs make organizations more innovative and more collaborative. En-
abling and investing in COINs is an enormously cost-efcient way to innovate
and get ahead of the competition, as DaimlerChryslers experience shows.
How? Organizations that make supporting COINs an integral part of their
culture become more innovative by creating a transparent innovation process
and making it possible to reward hidden innovators. People in organizations
that support COINs become more efcient, as well as effective at work-
ing together. Tim Berners-Lee was working at CERN, the European nuclear
physics consortium, when he began to create the World Wide Web. Although
the effort was far aeld of CERNs core competency, the organization let
him proceed and won considerable recognition for doing so. A company can
always decide later whether innovations t into the strategy and whether to
invest in them.
COINs make organizations more agile. Organizations that support COINS
are better equipped to react to market and technology changes. Daimler-
Chrysler showed that by leading the way in establishing the main online car
parts supplier marketplace. And UBS became one of the rst banks to have a
Web-based intranetat very low cost. (This is detailed in chapter 5.)
COINs infuse external knowledge into the organization. COIN members
have knowledge of what is going on in the marketplace, and they usually take
initiative in building up personal links with peers outside the organization. By
supporting COINs, organizations can improve and build on that knowledge.
The core COIN team at DaimlerChrysler was well aware of what General
Motors and Ford were up to, and was able to advise its own management to
join forces with those companies to establish an online car parts marketplace.
At UBS, knowledge of the ease of use and simplicity of Web technologies
COINs and Their Benefits
13
convinced the team to ignore the initial resistance and move forward with the
Bank Wide Web.
COINs uncover hidden business opportunities. Supporting COINs allows
a company to identify hidden business opportunities well before the com-
petition. As previously mentioned, at DaimlerChrysler, a COIN allowed the
company to participate in the largest online car parts marketplace before
its European and Asian competitors. At IBM, a small COIN of midlevel
managers was rst to spot the signicance of the Internet for the companys
business. Lobbying tirelessly, the COINmembers convinced top management
to put IBMs full weight behind e-businesswhich generated $40 billion in
revenue for the company in 1999.
1
COINs release synergies. At near-zero cost, Deloitte Consulting in Europe
was able to leverage existing resourcesboth people and infrastructureas
a COIN to create an entirely new service offering and generate lots of new
revenue. (This is detailed in chapter 5.)
COINs reduce costs and cut time to market. By supporting nimble and
exible COINs, organizations create a more transparent environment, which
leads to greater trust and, in turn, to reduced transaction costs. Time-to-market
for new inventions can be reduced substantially. In fact, COINs offer an
enormously cost-efcient way to invest in innovations. People normally join
a COIN not for immediate monetary reward, but because they are fascinated
by a problemand want to contribute to solving it. This was true for the Deloitte
Consulting practice, which not only was set up in a few months at no cost,
but alsoat the height of the dot.com crazeserved as a highly efcient and
low-cost retention device for highly marketable staff.
COINs help organizations locate their experts and reward hidden contrib-
utors. By supporting COINs, an organization can identify its internal sources
of tacit knowledge and innovation. Frequently, these people are not high up
in the ofcial hierarchy and are not recognized by management. If supported
appropriately, though, they offer tremendous value. By recognizing that value
and making the merits of these people more obvious, an organization not only
improves morale but also enhances organizational productivity and protabil-
ity. At DaimlerChrysler, the tremendous contributions by a group of people
fromoutside the ofcial project became obvious, thanks to the COINstruc-
ture of the project. Once on top managements radar screen, these people were
quickly promoted to ofcial leadership positions within the project and the
business units that grew out of the project. At UBS, it became obvious that
swarm creativity
14
the skills and tutoring of one particular programming wizard had been crucial
to the success of the project. He became one of the rst non-managers named
a vice president at the bank.
COINs lead to more secure organizations. The higher transparency and
the high-trust operating environment of COINs have signicant benets,
as transparent communication ow exposes security risks early on. This is
apparent in the examples of several COINs, such as the Internet Engineering
Task Force (IETF), the World Wide Web Consortium (W3C), and Linux
developers. They operate in complete transparency: everybody posting to a
mailing list in IETF, W3C, and Linux working groups knows that everyone
else can read their postings. Impostors, spammers, and anyone trying to use
the groups and their lists for any reason other than to benet the community are
immediately exposed, policed, andif unwilling to rectify their behavior
expelled.
These are some of the many quantitative and qualitative benets organiza-
tions can realize through COINs. Well explore themmore fully in subsequent
chapters. But why should individuals be willing to dedicate their time and col-
laborate in COINs? If organizations arent encouraging their people to work
in COINs, is it worth it? What happens if senior management doesnt see the
value?
COINs Benet Individuals
Even without initial managerial backing, and in the face of organizational
obstacles, there are still strong incentives for individuals to join edgling
COINs.
People initially join COINs because they are fascinated by the challenge
and care deeply about the goals of their community. Their primary currency of
reward is peer recognition. Programmers participate in open-source software
projects such as Linux because they want to further the development of their
product, and because they want to build a reputation as Web or Linux experts.
Consultants joined Deloittes e.Xpert group because they wanted to become
e-business experts (e.Xperts) and be recognized as such. DaimlerChryslers
project was started because team members wanted to become recognized as
e-procurement experts. The Bank Wide Web at UBS was started because the
COINs and Their Benefits
15
core team believed in the future of the Web and wanted to apply it to daily
business tasks.
There are more tangible benets for individuals, as well.
COINs help individuals build a wider network that bridges structural
holes and get direct access to business-critical information. COINmembers
get to know deeply committed people with shared values and beliefs whom
they might not otherwise have met. This allows COIN members to build up
an invaluable network of people to draw on in professional situationsand
ll in the structural holes in their existing personal networks. The members
of the Deloitte Consulting e.Xpert COIN tapped into the skill base of their
fellow e.Xperts to complete successfully their own e-business engagements.
Further, people participating in COINs discover links and synergies they
never knew existed. Members of the e.Xpert network at Deloitte Consulting
were tipping each other off on new business leads and e-business project
opportunities well before requests came in through ofcial channels. This
meant that their utilizationone of the primary measures of success for a
consultanton average was substantially higher than that of the rest of the
Deloitte consultants.
COIN members build personal relationships with leaders in their eld.
People participating in COINs learn about experts in specic problem do-
mains. They have better opportunities to leverage the skills and knowledge
of othersfar beyond those normally advertised in the corporate yellow-page
directories. For example, the COINmembers in DaimlerChrysler got to know
luminaries of Internet commerce fromleading software vendors such as SAP,
CommerceOne, Manugistics, i2, and Ariba. This means that in their future
activitieseven outside of the specic projectindividual COIN members
can call on some of the worlds most highly valued experts in leading-edge
technologies.
COINmembers learnnewskills, become experts, andoftenndthemselves
promotedrapidly. Over time, people participating in COINs themselves
become experts. They become more exible in adapting to market and
technology changes; they obtain access to innovations that are under way in
their organization. This creates new opportunities for personal development
and improves prospects for personal advancement. For example, people
working on the Bank Wide Web project at UBS became early Webmasters
and Internet experts. They had an edge well before the dot.comhype, and later
swarm creativity
16
found themselves promoted rapidly within the bank or lured to high-paying
positions with outside consulting rms.
Transforming Organizations to Support COINs
The blood system energizes the human body, but the human body also needs
the nervous system to steer and supervise its vital functions. Similarly, while
a companys lifeblood ows through its business processes, the long-term
health of the company depends in large part on its knowledge ow(gure 1.1).
Over the past two decades, businesses have largely focused on stream-
lining structured business processes. Today, the challenge is to optimize the
ow of knowledge, streamlining unstructured, knowledge-intensive innova-
tion processes, and turning organizations into COINs. By visualizing the ow
Figure 1.1. Business processes are the blood system; knowledge flow is the nervous
system.
COINs and Their Benefits
17
of knowledge, making it transparent, and optimizing its course, organizations
and individuals become more creative, innovative, and responsive to change.
This is one of the keys to success in the newcentury. But while the importance
of continuously optimizing and ne-tuning business processes is universally
recognized, the importance of redesigning and optimizing knowledge ows
is still widely underrecognized.
Organizations can successfully promote COINs by giving up central
control in favor of self-organization in swarmcreativity, developing an ethical
code, and setting up a social network connected by hubs of trust (which
youll also learn about in subsequent chapters). The lessons learned for
COINs and cyberteams apply to every company and organization in daily
business life. Companies that have already harnessed the principles of COINs
have beneted by successfully creating new, innovative products or making
existing processes more exible, efcient, and agile.
COINs will be the foundation of virtual teamworking for tomorrows
increasingly virtual global enterprises. Innovation is crucial to the long-term
success of anorganization, andCOINs are the best engines todrive innovation.
To understand how and why, and to see how results from the companies
presented in this book relate to your company, you need to understand swarm
creativity. That is the subject of the next chapter.
This page intentionally left blank
CHAPTER2
Collaborative Innovation through Swarm Creativity
It is an odd word, not one that youll nd in the parlance of most organizations,
business or otherwise. But the idea of swarm, especially in the context of
creativity, promises to become more and more familiar to those seeking to
stay on the cutting edge of innovation in the new century.
In July 2003, Boston experienced the power of swarming. Flash mobs had
already burst into applause on the mezzanine of New Yorks Grand Hyatt,
and they had whirled like dervishes in San Francisco. On Thursday, July 31,
2003, they struck in Boston. As Harvard bookstore employees watched, a few
hundred people crowded into the greeting card section, holding instructions
on what to say and how to act. They were looking for a card for their friend
Bill. He lives in New York. Ten minutes later, the mob broke into applause,
and then dispersed. Bill fromNewYork is credited with creating ash mobs.
The gatherings are coordinated through Web sites and chain e-mails, with the
point of being pointless. While the rst ash mobs strove to stay silly, they
have since grown from an Internet curiosity to an increasingly widespread
part of modern Web society.
Even politicians like 2004s U.S. Democratic presidential candidate How-
ard Dean have been drawn into the ash mob phenomenon to organize
their political campaigns. Dean relied heavily on the World Wide Web to
mobilize his grassroots supporters, to collect campaign money, and to ne-
tune his political message by collecting direct feedback fromsupporters on his
19
swarm creativity
20
Weblog. John Kerry, the 2004 Democratic presidential nominee, continued
when Dean dropped out, collecting a large part of his more than $300 million
campaign donations fromgrassroots supporters using his Web site. The Kerry
campaign learned the lesson fromDean. Thus, irrespective of what you might
have thought of Deans politics, no one can deny the power unleashed by his
campaigns swarm.
Why should this matter to business executives?
Swarm Intelligence for Social Insects
Flash mobs are striking examples of how the Internet can be tapped to
coordinate swarming behavior, a concept popularized by computer scientist
Eric Bonabeau. Having studied the amazing world of swarm intelligence, the
collective intelligence of social insect colonies, Bonabeau is now applying
his insights to human interactions and computer technology.
Individually, one insect may not be capable of much; collectively, social
insects are capable of achieving great things. They build and defend nests,
forage for food, take care of the brood, divide labor, form bridges, and much
more. Look at a single ant, and you might not think it is behaving in synchrony
with the rest of the colony. But we sometimes observe ant highways
impressive columns of ants that can run from tens to hundreds of meters.
They are highly coordinated forms of collective behavior.
Swarm intelligence in social insects is based on self-organization; no one
is in charge, but social insects successfully solve complex tasks. According
to Bonabeau, self-organization has four properties:
1
1. Positive feedback reinforces desired behavior, such as when a bee
recruits other bees to help exploit a food source
2. Negative feedback counterbalances positive feedback, such as when
bees overcrowd a food source, which stops them from exploring it
3. Amplication of randomness leads to positive reinforcement, such as
when bees that get lost trying to locate a known food source discover
new food sources
4. Amplication of interactivity has a positive outcome, that is, when
insects make positive use of the results of their own activities as well
as those from the activities of other insects.
Collaborative Innovation through Swarm Creativity
21
Social insects combine these four properties into predened patterns, which
have evolved over time, to efciently accomplish a given task. For example,
colonies of ants can collectively nd the nearest and richest food source even
if no individual ant knows its location.
Errors and randomness contribute very strongly to the success of social
insects by enabling them to discover, explore, and exploit. Errors feed self-
organization, creating exibility so the colony can adapt to a changing envi-
ronment with robustness, ensuring thateven when one or more individuals
failthe group can still perform its tasks.
Bonabeau reasons that our world is becoming so complex that it cannot
be comprehended by any single human being. Swarm intelligence offers an
alternative way of designing intelligent systems in which autonomy, emer-
gence, and the ability to distribute tasks replace control, preprogramming, and
centralization. During most of our history, we human beings have suffered
from a centralized mindset: we like to assign the coordination of activities
to a central command. The self-organization of social insects, through di-
rect and indirect interactions, is a very different way of performing complex
tasksand it closely resembles the behavior of collaborative innovation net-
work (COIN) team members. Open source developers, for instance, exhibit
behavioral patterns similar to those in an ant colony. While the behavior of
the individual programmer might appear random, open source developers are,
like ants, self-organizedtobuildimpressive software systems, directedbylead
developers such as Linus Torvalds (the open source ant queen), whoimpress
their distinctive brand and avor on their colony of software developers.
The crucial point here is that in social insect colonies, as in COINs,
there is no individual giving marching orders. Neither the ant queen nor the
lead developer of an open source team directs the individual behavior of the
individual.
The obvious advantages in accomplishing complex tasks through swarm
intelligence include no central control, errors being okay, exibility, robust-
ness, and self-repair. If a decentralized, self-organizing system takes over,
though, how should the individual behave so that the network performs ap-
propriately at the system-wide level? It is difcult for conventional managers
to accept the idea that solutions are emergent rather than predened and pre-
programmed. Most managers would rather live with a problem they cannot
solve than with a solution they do not fully understand or control. The attitude
of many managers to COINs reects this view; managers resent giving up
swarm creativity
22
control in exchange for self-organization, increased agility, and increased
exibility of their organizations.
Watched individually, members of a COIN may appear, like individual
ants, to behave erratically. Like an ant colony, though, the entire COIN op-
erates as a highly efcient self-organizing community. Its members commu-
nicate in patterns structured as small-world and scale-free networks (which
we will learn about in more detail in chapter 4). They share a common code
of ethics that forms a sort of collaborative bond based on similar expertise
and shared goals. In the appendixes, we will see how to make use of the
e-mail pheromone trail of COIN members (much like the chemical trail
ants use to communicate and collaborate) to spot COINs and to optimize
their performance.
Lessons learned from the amazing world of swarm intelligence of social
insects also apply to how members of COINs innovate. The ant or bee queen
is not served because she orders her insects to do so, but because evolution
has taught the insects that protection of the queen means protection of their
gene pool for survival. The same is true for the people working together in
a COIN. People working with the innovator are not working for her or him
because they have been ordered to do so, but because they want the innovation
to succeed. They all share the same vision and goals (in a sense, the same
genes); they want to succeed, and they want to see their innovation spread
and be accepted by the outside world.
What Is SwarmCreativity?
Perhaps nothing expresses the main motivation for swarm creativity better
than something I read on a napkin in a San Francisco restaurant: If you and
I swap a dollar, you and I still each have a dollar. If you and I swap an idea,
you and I have two ideas each. By openly sharing ideas and work, a teams
creative output is exponentially more than the sum of the creative outputs of
all the individual team members. While swarm intelligence is based on equal
sharing of information, swarm creativity is founded on sharing ideas openly.
COIN members share their common vision. They share all their ndings and
results of their common work, and they also share the credit for the results of
their work. COINs collaborating in swarm creativity can achieve awesome
results.
Collaborative Innovation through Swarm Creativity
23
One of the most amazing examples of swarm creativity I know of is a
virtual collaboration among three jazz composers and musicians, who play
together at the same location but compose and record the same piece inparallel
by coordinating over the Internet. Without ever getting together physically,
they jointly composed and recorded an opera called Popes.
The three composers who succeeded in this team effortDavid Hoenigs-
berg, Douglas Grannell, and Gareth Dylan Smithmet several years ago
and spent considerable time playing and improvising together. They discov-
ered their mutual compatibility. They shared common behavioral and musical
patterns, a joint way to tackle a musical task, or, as Hoenigsberg calls it, a
common sound world. This is familiar for accomplished jazz musicians,
who get together to play for hours on end without written melodies. They
operate as a perfect team, where a theme is taken up and passed from one
instrument to the other, with team members switching from playing in the
background to getting the solo part.
Hoenigsberg, Grannell, and Smith extended the same concept into the
cyberworld, virtually improvising together over long distances when they
composed their joint opera. The three men were immersed in the same virtual
world, collaborating as mutually aligned musicians, using the same musical
and behavioral patterns. They produced an end product in which the sum
is larger than its partsa product none of them could have produced indi-
vidually. In creating Popes, Hoenigsberg, Grannell, and Smith provided a
perfect blueprint for a self-organizing, creative team innovating together and
collaborating by sharing a common behavioral code.
Its notable that the three composers were not paid to do their work.
They could not expect any immediate nancial reward or immediate fame;
moreover, their style of music is appreciated only by a relatively small group
of dedicated lovers of modernist classical harmonic music. Nevertheless, they
all shared the same musical goal, and they formed a highly motivated team
that overcame considerable obstacles in creating an opera.
Swarm creativity needs enough leeway to ower. Team members want
a risk-free work environment where they get mutual emotional support.
Hoenigsberg, Grannell, and Smith work together as a team of equals, char-
acterized by high mutual respect for each others capabilities. When I inter-
viewed them (individually), each spoke of the others with the greatest respect
and tolerance for their strengths and weaknesses. Their relationship is based
on a high level of trust, which has been built up by playing together over
swarm creativity
24
the years. The success of the long-distance Popes collaboration strengthened
that trust.
Team members withhold no information. The free ow of ideas and
thoughts is essential to the success of creative teams. While Hoenigsberg,
Grannell, and Smith have a clear split of musical responsibilities, with a clear
picture of which components each will compose, they stay in close contact.
They exchange e-mails and phone calls frequently, bouncing inspirations
and melodies back and forth and gauging each others reactions to new
ideas. This is not a process where there is always unanimous agreement.
Rather, disagreements are raised openly and discussed. From this conict,
new creative inspirations arise.
This composer team has no external coordinators. It is entirely self-
organizing and self-selecting. Roles and responsibilities of each member are
clear to all, with no need for lengthy coordination meetings. There is no
ofcially appointed leader, and Hoenigsberg has assumed a coordinating role
only because, in his words, he sort of naturally inherited the function. He
sees himself not as an authoritative boss, but as a servant of his team who
does his best to help each team member succeed in his chosen task.
The Popes project, then, illustrates excellently the various facets of swarm
creativity. However, it is only a three-person innovation team. There are
COINs with tens of thousands of members engaged in swarm creativity. An
ideal example is Wikipedia: it is the ongoing result of the creative output of
more than 100,000 volunteer authors and editors.
WikipediaSwarmCreativity Thriving Online
Wikipedia, the online encyclopedia, thrives on swarm creativity. It is the
collaborative, nonprot competitor to commercial online encyclopedias such
as Microsoft Encarta and Encyclopaedia Britannica.
2
Wikipedia was founded in 2001 by Larry Sanger, a philosophy lecturer at
Ohio State University, and Jimmy Wales, an Internet entrepreneur. While
Encarta and Encyclopaedia Britannica are maintained by a staff of paid
journalists, researchers, and scientists, Wikipedias more than 400,000 entries
are updated by thousands of volunteers. Wikipedia makes no distinction
between authors and readers; any reader can change an encyclopedia entry at
Collaborative Innovation through Swarm Creativity
25
any time. The current directory of Wikipedians, as the volunteer contributors
call themselves, lists more than 100,000 names, extended by an unknown
number of anonymous authors.
Wikipedia entries are surprisingly accurate and complete because they can
be corrected and updated. Authors are asked to maintain a neutral viewpoint
(as well as to be bold but stay cool and to be nice to newcomers)
in order to keep the contents objective. A small group of trusted regular
authors and editors are appointed as administrators because not everyone
follows the guidelines and etiquette, or Wikiquette. The strongest measure
of censorship is that this small group can temporarily turn off editing of an
entry for the rest of the world.
3
Thus far, Wikipedia has worked remarkably well, based on a policy of
nearly complete freedom to edit. The Wikiquette asks each Wikipedian to
treat her or his fellowcommunity members with respect. Wikipedia is entirely
transparent in how text for controversial entries is fought over by different
factions in the maintainer team. Alog for each entry tracks the change history.
Each entry also has a separate discussion page, where disagreeing authors are
asked to engage in polite dialogue and where people should be prepared to
apologize. An important part of the Wikiquette describes how to give praise
when praise is due by making friendly entries on a users page, by listing work
under the category great editing in progress, and by nominating entries as
the days featured article.
Wikipedia is not the only COIN sharing the fruits of its collaborative
endeavors at no cost. The open source movement is one of the best examples
of a network of COINs, which innovate in swarmintelligence by contributing
their intellectual property to the world and giving away their software source
code for free.
Open Source SoftwareThe Advantage of SwarmCreativity
One of the most compelling arguments for operating in swarm creativity is
the astonishing rise of the Linux operating system. Finnish computer science
student Linus Torvalds was certainly unaware when he began his termproject
to develop a UNIX-based operating system for the personal computer (PC)
that Linux would becomein just a little over 10 yearsthe main competitor
swarm creativity
26
to Microsofts dominance of the small server and desktop operating system
(OS) market. The search engine giant Google uses Linux to power its farm
of thousands of Web servers.
Linux has become the most popular alternative OS to Windowsnot just
because it is available for free, but also because anybody can make changes
to it. The entire source code of Linux is publicly available, and anyone adept
at the C programming language can extend the code. Torvalds asks only that
improvements to Linux be shared; he and a group of volunteers check them
and integrate the best into future versions of Linux.
What Torvalds created is open source software, which comes in many
avors but shares certain characteristics. One is that the source codethe
underlying instructions that determine how a program operatesis public,
so users can tamper with it; another is that new code is usually contributed
by volunteers. So far, most open source projects are concentrated in areas of
infrastructure softwarethe code that runs the core activities of a computer
or a network. This extends fromthe basic operating systemto the middleware
layers of software needed to run particular applications.
The success of open source software projects illustrates the power of
COINs and relates directly to some of the benets introduced in chapter 1.
Key to the success of Linux, for instance, has been total transparency and a set
of consistent rules among thousands of programmers involved. Advocates of
open source also claimwith considerable justicationthat their software
is of a higher quality than other software. Because there are so many users and
developers of open source software products, there wil be no deepthat
is, hard to ndbugs, or software glitches. This has been veried by several
research groups.
4
In the case of closed source software, users usually update only after
software programmers have xed reported bugs. A new version is available
only after discrete periods of time, during which private programmers work to
x the last batch of faults. For open source software, this type of feedback and
updating happens continuously. Open source software has a faster develop-
ment and x cycle compared to proprietary systems, and it also benets from
more user feedback. The same factors also help open source software develop
stronger security protection, making open source software more secure than
proprietary software.
The conclusion: open source software is developed faster, better, more
securely, and at lower cost than closed source software. Teams of open source
Collaborative Innovation through Swarm Creativity
27
software developers operate as near-ideal COINs, coordinating their creative
efforts through swarmintelligence. Their products are at least as good as those
produced by conventional closed source programmer teams in conventional
companies. At the same time, open source programmers also operate as a
meritocracy, where the intellectual and emotional capabilities of individuals
dene their position in the open source development community.
IBM is one company that has discovered the advantages of COINs in
combinationwithopensource. The computer giant has investedmore than$60
million into the open source tool set Eclipse, which has become the leading
Java software development environment, attracting hundreds of vendors and
millions of users. In fact, growth and adoption of Eclipse have been so rapid
that companies such as Ericsson are evaluating it as a single, corporate-wide
development platform to cut costs and boost efciency.
In the beginning of 2004, IBMcompletely released its control over Eclipse
and turned it over to an independent foundation. IBM gets to appoint just one
of 10 members of the foundations board of directors, but it continues to invest
substantially in Eclipse by employing more than 50 Eclipse core developers
with committer statusthat is, people who form the innermost core of the
Eclipse open source developer community, owning the rights to make changes
directly to the source code.
Why does IBM invest in Eclipse under an open source license? I met with
Erich Gamma, one of the lead developers of Eclipse, in his IBM ofce in
Switzerland. First, he told me that IBM recognized that it cannot build all the
software tools itself, that it neededa platformtohelpintegrate different tools so
that partners can contribute and extend the tool set. Obviously, such a platform
also helps IBMalign its internal tooling efforts; having a widely used software
platform distributed under an open source license guarantees high quality, be-
cause it is tested and maintained by millions of programmers who discover,
report, and x bugs continuously. Second, IBMsells complementary software
tools based on Eclipse, and having Eclipse available for free on almost any
platform provides IBM with a huge potential customer base for these tools.
Third, IBM reaps lots of fame, goodwill, and what Gamma calls the cool
factor by being seen as a signicant contributor of the best Java develop-
ment environment to the public domain. And IBM is using Eclipse to develop
some of its own commercial software packages, such as Lotus Workplace.
The Eclipse project provides a prime example of a COIN that has had a
stellar success based on principles of meritocracy, transparency, and openness
swarm creativity
28
garnering numerous awards as the best Java product and Java development
tool (see sidebar 2.1). Even more important, users have cast their vote,
downloading the Eclipse tool more than 18 million times since its inception.
Intrinsically motivated and self-organizing into different roles of work,
open source developers perfectly demonstrate the advantages of swarm in-
telligence (see sidebar 2.2). They produce results of superior quality at much
lower cost than do commercial software developers. Working to develop new
Sidebar 2.1
The Eclipse Development TeamOperates as a True Knowledge Network
The core innovation network for the Eclipse technology project is surrounded
by learning and interest networks. Specifically, former Object Technology
International development labs in Ottawa, Toronto, Portland, Minneapolis, Saint
Nazaire, and Zurich (since acquired by IBM) comprise the networks backbone.
Each of these labs hosts a small team of developers led by Eclipse project leaders;
the leaders act as hubs of trust linking the core teams into a globally active COIN.
These leaders have collaborated closely for many years and have established high
mutual trust. The trust is then transferred to programmers working on teams at
different locations.
Erich Gamma---an Eclipse lead developer who works at IBM in Switzerland---is
one such hub of trust. He emphasizes meritocracy, transparency, and openness to
contributions from everyone as the three basic principles governing the Eclipse teams
work ethic. IBM employees receive no preferential treatment. New programmers,
whom Gamma hires for his team at IBM Research in Zurich, must be confirmed
and voted on by their Eclipse project peers, independent of company affiliation.
A new hire begins as a debugger, writing software patches to fix bugs.
Debuggers work their way up to become developers, who actively participate
in newsgroups and mailing lists. Next, they can be voted in as committers,
assuming they have proven their worth and gained a reputation as trustworthy
programmers; committers gain write access to the source code repository and
voting rights for new committers.
The project is managed by a project management committee (PMC), which
today has three active members (the size has varied). New PMC members are
admitted only after unanimous approval of all active PMC members (including
Erich). In this way, decision processes regarding project management and
tool development strategy are made transparent to the entire community.
Membership in the Eclipse foundation is open to everyone.
Collaborative Innovation through Swarm Creativity
29
Sidebar 2.2
What Motivates an Open Source Developer to Be a COINMember?
Over a beer, Stefano Mazzocchi---the inventor of the open source Apache
Cocoon framework---explained to me what inspires him. Theres a three-
level pyramid that motivates open source developers. At the lowest level, open
source programmers want to boost their ego by gaining a reputation as supreme
programmers. At the next level up, theres the desire to gain new knowledge. At
the highest level of motivation is the fun factor, of enjoying what they are doing.
Additionally, an online survey of members of the Linux developer community
details three categories of motivation.
1
1. Collective motives include identification of the developers with the goals of
their community and the strong desire to help make those goals come true.
2. Peer recognition motives, not as important, include the wish to work together
with friends and colleagues---and for those people to rank you highly.
3. Direct reward motives include learning new skills, making more money, or
gaining new friends. Open source developers are motivated primarily not by
money; having fun and learning ranked highest in the survey.
1. See www.psychologie.uni-kiel.de/linux-study/.
innovative products by massive collaborative creativity, they are powered by
the self-organizing tenets of swarm intelligence: reinforcement of positive
and negative feedback, and amplication of randomness and interactivity.
Its a power that businesses can harness by supporting COINs.
More Examples of SwarmCreativity
I experienced swarm creativity rsthand when we created the Interoper-
ability Service Interface (ISI) for Union Bank of Switzerland. We began
by using public domain software. The project was running on a shoestring
budget, driven by the dedication and commitment of the team members;
these members had complementary talents, combining programming, system
architecture, database, algorithms, management, and sales skills. The primary
motivation for teamwork was clearly not nancial, but to get the ISI system
swarm creativity
30
up and running, and membership was highly self-selecting. After all, while
this was an opportunity to learn about the latest software technologies, it was a
high-risk project and a huge technical challenge. Within the team, we worked
in a high-trust environment with internal transparency. It was okay to have
crazy ideas, which the entire team would discuss. In a few weeks, we were
able to build a successful prototype system at very low cost, and ISI is still up
and running. (The banks COIN is described in greater detail in chapter 5.)
DaimlerChryslers e-extended Enterprise (e
3
) project, introduced briey
in chapter 1 (and detailed in chapter 5), also illustrates the properties and
advantages of swarm intelligence in a COIN with obvious commercial goals.
The project was a product of the e-business economy of the late 1990s, and
so merely being part of this new and revolutionary undertaking was highly
motivating for project team members. It was unbelievably exciting to partic-
ipate in this journey into uncharted territory, at the leading edge of the eld,
at barely controllable speed. Nobody knew what the next day would bring.
Unexpected developments and alliance opportunities popped up constantly.
In those spring months of 2000, the entire team worked at 200 percent of its
capacity, energized by the vision of its executive vice president, who let the
team members take the reins and set the day-to-day directions themselves.
Clearly, the motivation here was not opportunities for raises and promo-
tions, but the enjoyment of participating in a total overhaul and a melding to-
gether of the procurement processes of Daimler and Chrysler. This excitement
was enough to overcome widespread initial skepticism of the management of
the procurement and supply departments at both rms involved in the merger.
Department heads on both sides of the Atlantic initially were reluctant even to
share simple information such as the head counts of their business units, but
their progressive buy-in into the shared vision of creating something totally
new resulted in one unied COIN team. The outside torrent of e-business
also created an environment of creative chaos and uncertainty that greatly
increased the readiness of executives to accept change.
The project team worked together as a real collaboration network. Once
the initial reluctance to share knowledge was overcome and mutual trust
had been developed, information was made transparent to all team members.
Although the team was physically split between locations in Stuttgart and
Detroit, team identity was quite strong. Having monthly face-to-face work-
shops alternately in Germany and the United States helped build a shared
identity and a common code of ethics; in those close interactions, strengths
Collaborative Innovation through Swarm Creativity
31
and weaknesses of each team member became obvious to the entire team,
leading to an open and honest communication style. Team members, regard-
less of their rank in the company, were not afraid to speak out to the entire
group. Complex problems were broken down into tasks and assigned to team
members on both sides of the Atlantic. Thanks to a shared common under-
standing, multifaceted issues such as the development of a new supply chain
model or the evaluation of supply chain automation software tools could be
solved efciently by team members working in parallel in both locations.
Membership in the team was self-selecting. Some team members who
were initially appointed by senior management pulled out voluntarily, while
others who initially were just seconded for a certain presentation or a specic
task became increasingly engaged and asked to be ofcially transferred into
the project, subsequently becoming mainstays of the team.
In the end, the e
3
project was a big success by most measures of project
management. Not only were most deliverables on time, but they were also
developed efciently and at reasonable costs. Newprocurement processes and
software solutions were successfully deployed, and the entire project more
than paid for itself in the rst year of operations. Project team members
moved on to assume prominent roles within DaimlerChryslers procurement
organization and were even chosen for leadership positions with outside rms.
Again, the business success of this project is undisputable: the company
spent slightly more than $10 million to optimize its procurement function and
saved more than $50 million in the rst year of operation. And it succeeded
because of collaborative innovation.
Why Collaborative Innovation Is So Powerful
What starts a COIN, and what triggers successful innovation? Is it the phase
of the moon, or just sheer luck, that is responsible for creative inspirations
and the emergence of groundbreaking new ideas? The crucial point about
innovations developed in COINs is that they are open and disruptive.
Harvard professor Clayton Christensen distinguishes between sustaining
and disruptive innovation.
5
Sustaining innovations do existing things in a
novel and better way. For example, Wal-Mart revolutionized mass retailing by
using information technologies in a much more massive and better-integrated
way than its competitors. It optimized its supply chain by pioneering concepts
swarm creativity
32
like vendor-managed inventory and automated demand forecasting and re-
plenishment, beating its competitors and becoming the number one Fortune
500 company globally, as well as one of the most successful enterprises by
many measures. But the innovation of Wal-Mart consisted of doing the
same things that its competitors did, only better, much more efciently. For
example, Wal-Mart used available information technologies more thoroughly
and consistently than its competitors by being one of the rst companies to
let its suppliers manage shelf space in Wal-Marts own stores.
On the other hand, disruptive innovations such as the World Wide Web
and Linux, the airplane, and the railroad totally changed the way in which
something was done.
6
Disruptive innovations are normally conceived outside
of established organizations or, if they come out of organizations, they are
developed without organizational blessing and are not aligned with the ofcial
organizational vision and goals. They have the potential to turn established or-
ganizational norms upside down and to redene the way people work and live.
Academics such as Christensen and Eric von Hippel have analyzed how
disruptive innovations succeed.
7
Their main insight is that disruptive innova-
tions rarely become embedded into large enterprises. While they might initiate
withinlarge organizations, theysucceedoutside the formal hierarchies of huge
enterprises. In fact, the teams that take new inventions outside of the large
corporations to start new companies frequently initiate as stealth COINs
inside large corporations.
In the past, the centrally funded research labs of large corporations were
responsible for many disruptive innovations. AT&T Bell Labs (now part of
Lucent Technologies), IBM Researchs labs, and the Xerox Palo Alto Re-
search Center (PARC) produced the transistor, speech recognition, copper
chip technology, the scanning tunneling microscope, and the Ethernet; they
all fundamentally changed the way a certain product or process operated. But
the closed innovation model, in which companies fund these central labs to
develop new technology and products, is no longer generally valid. What
Harvard professor Henry Chesbrough calls the open innovation model is the
new way in which individuals, small companies, and innovation networks
create innovations.
8
The obsolescence of the closed innovation paradigmhas been accelerated
by different factors. The explosive growth of a large base of skilled, mobile
workers has created an auction market that allows startups and entrepreneurs
to hire the best and the brightest away from larger companies, which
Collaborative Innovation through Swarm Creativity
33
threatens the bigger rms ability to sustain research and development (R&D)
investment. Between 1980 and 2000, U.S. venture capital skyrocketed from
roughly $700 million to over $80 billion. Ready access to this capital,
in combination with lucrative stock option packages, attracted many key
lab personnel from large companies to startups, eroding the knowledge
bases of some larger companies. The availability of venture capital, worker
mobility, and global communication means that research employees no longer
have to keep their ideas on the shelf while they wait for internal approval
to explore them. Instead, employers rules permitting, they can commercialize
those ideas independently. External suppliers are becoming more procient
at delivering products that are just as good, if not better, than what an internal
corporate R&D lab can developa trend that levels the playing ground
betweenlarge rms andtheir smaller competitors. It allows large rms tobring
new products to market faster, while giving smaller rivals the opportunity to
overtake larger companies. To take advantage of the changes brought by those
factors, companies must move to the open innovation model that leverages
the knowledge base of COINs. If they dont, COIN members will leave and
start their businesses elsewhere!
Disruptive grassroots innovations by COINs, such as the Internet, the
Web, and Linux, grew out of research labs and universities with little or
no organizational blessing and with minimal budgets. The main drivers of
these innovations were not the compensation-and-reward systems of large
corporations, but the dedication and commitment of the researchers and
innovators. IBM, Intel, Microsoft, and other companies have learned this
lesson and are successfully applying the same principles of open innovation
to their own internal research processes. They have intensied collaboration
with university researchers and small knowledge companies, and they take
newinnovations to market much earlier than in the past. Amore collaborative
culture combined with open communication has led to increased innovation,
ultimately resulting in radically newrevenue opportunities for companies that
embrace the open innovation paradigm.
Open disruptive innovation is not restricted to the high-technology area.
For instance, the people spearheading snowboarding are typical open dis-
ruptive innovators (see sidebar 2.3). They are pioneersinitially frowned
upon by the majority. They are not afraid of riding off-slope, and they are
developing their own solutions to manage their journey into new territory
until their inventions become an accepted part of daily life. The commonality
swarm creativity
34
Sidebar 2.3
Snowboarding as a Disruptive Innovation
The meteoric rise of snowboarding in just over 30 years is an excellent showcase
for open disruptive innovation in a non--high technology setting.
The sports earliest innovator was Jack Burchett, commonly credited with
building the first snowboard in 1929 when he cut out a plank of plywood and
tried to tie his feet to the plank with some clothesline and horse reins. Sherman
Poppen, a chemical gas engineer in Michigan, took up the idea 35 years later,
inventing the snurfer for his daughters by securing two skis together and
putting a rope at the nose that the rider could hold onto to keep the board more
stable. The snurfer was reasonably successful, and Poppen started a company
to market his invention. But the snurfer was only seen as a toy for kids. It wasnt
until the 1970s that snowboarding really took off. Dimitrije Milovich, a college
student in upstate New York, worked as a waiter and was sliding down hills
on cafeteria trays. He and other enthusiasts began to build snowboards that
combined the concepts of surfboards and skis.
In 1982 the first U.S. National Snowboard race was held in Suicide Six, outside
Woodstock, Vermont. Three years later, although a mere 39 of the approximately
600 U.S. ski resorts allowed snowboards, the craze had spread. St. Moritz,
Switzerland, hosted the first snowboarding world championship in 1987. By 1994,
snowboarding was an Olympic sport.
The early snowboarding pioneers wanted nothing to do with the ski establish-
ment, and the feeling was mutual: most skiers frowned upon the brave souls who
rode their own devices and the first snowboards. Today, though, snowboarding
between the inventors of snowboarding and high-technology pioneers like
Tim Berners-Lee is their combination of capabilities as team players with
creative intelligence.
Fritz Bircher, a professor of electrical engineering and computer science at
the University of Applied Sciences of Burgdorf, in the Swiss canton of Bern, is
an example of the kind of creator who combines emotional intelligence with
creativityandproves anideal inuencer for a COIN. Infact, he is the archetype
of a creatorconstantly on the lookout for new, interesting applications for
his latest invention, the MobileJet, an inexpensive inkjet printer on wheels
that can print arbitrarily large areas on almost any surface. Bircher started
the MobileJet project in 2000, when professors at another Swiss vocational
college asked one of their Burgdorf colleagues for help. Now, as we met, he
Collaborative Innovation through Swarm Creativity
35
is a decidedly mainstream activity, and snowboarders are as welcome as skiers on
the slopes.
Sherman Poppens Snurfer.
1960s Brunswick Snurfer ad
courtesy of Brunswick.
was getting a call from an investor who was interested in commercializing
his invention. But Birchers motivation is to design and build new things;
making money on his invention is of secondary importance.
Bircher is currently working with another Burgdorf professor, Reinhold
Krause, who is an expert in mechanical engineering. They have assembled
a team of about a dozen research assistants to bring their ideas to fruition.
Bircher talks constantly about how his teams members come up with solu-
tions by working together; in this aspect, he is a typical leader of a COIN,
where people work together in a meritocratic environment driven by a com-
mon vision and common goalsnot for short-termprot. Krause and Bircher
form the perfect core team of a COIN. Both are driven by high intellectual
curiosity, and their motivation to innovate is denitively not nancial. By
swarm creativity
36
fostering an open collaborative and meritocratic culture in which credit is
given where credit is due, results are produced not because people are paid to
deliver them, but because people care about what they are doing. This leads to
a continuous streamof newinnovations. For instance, the MobileJet printer is
being extended to work as an inexpensive and efcient Braille printer. Krause
and Bircher are also talking to artists about using the MobileJet to produce
large-scale works of art on concrete and tarmac surfaces. These two men have
rmly embraced the open innovation model, and they collaborate and share
their results with other innovators at universities and in industry research labs.
Krause and Bircher made headlines with another invention a few years
ago. Their global positioning system (GPS)-controlled lawnmower, with
autopilot, cut grass to an accuracy of two centimeters. This invention led
to a whole list of innovations: building the GPS system into huge combine
harvesters; using the same technology in a maintenance robot to measure and
maintain Swiss railway tracks; and adapting the technology to road rollers,
increasing the efciency and more evenly compressing asphalt coverings in
new road construction.
Krause and Bircher are creators and gurus exhibiting the right mix of
creativity, charisma, and collaborative skills. They excite other people about
their innovations. There are other types of inuencers who are crucial bringing
a new innovation over the tipping point to success. Lets look at one of the
most famous innovators of all timeLeonardo da Vinci.
Innovation by Collaboration
In addition to being one of the most creative innovators ever, Leonardo da
Vinci was an excellent teamplayer. His talent for dealing with other people in
difcult situations was legendary. His biographer, Giorgio Vasari, describes
how, when da Vinci was still an apprentice with the famous painter Andrea
del Verrocchio, his depiction of an angelic gure was so masterful that
Verrocchio considered giving up painting. Da Vinci replied that it was the
greatest compliment to the master that the student should exceed the masters
ability.
9
Later in life, da Vinci built up a community around him that included
renowned mathematician Luca Pacioli, as well as devoted younger artists
such as Andrea Salai and Francesco Melzi. Vasari wrote that da Vincis
Collaborative Innovation through Swarm Creativity
37
generosity was so great that he sheltered and fed all his friends, rich and
poor alike, provided they possessed talent and ability.
People like Leonardo da Vinci are lifelong learners, swarming together
with other intellectual giants. Pacioli, for example, not only invented double
accounting, but also worked with da Vinci on his engineering exploits. Da
Vinci purchased a copy of Paciolis famous book on Euclidian geometry
before the two ever met, and he drew the gures in their joint work On
Divina proportione, a book in which they studied the golden ratio and
related Euclidean theorems. Later, da Vinci and Pacioli taught each other in
a mutually enriching learning and tutoring relationship, with da Vinci clearly
having the sponsoring role.
Leonardo da Vinci is arguably the most famous inventor ever. But it makes
no sense to look at innovators in isolation. We need to study both the creative
genius and the people that surround him. On the one hand, the innovator needs
to be practical and down-to-earth enough to be understood by his surround-
ings; innovators must be able to position their inventions in their environment
and successfully convince others of their value to create sustainable progress.
Thats what da Vinci was able to do with his mechanical engineering and mil-
itary inventions and his paintings, frescoes, and statues. On the other hand,
the innovator must also get along with the people surrounding him. Each in-
novation needs a group of early adaptors and rapid followers that understand
the genius innovator and his innovative ideas. They form the beehive, or ant
colony, that acts as bridge and translator to spread the concept. At the core of
the COINare the charismatic, inspirational, and creative thought leaders who
are enough in synch with their time and environment for their innovations
to be recognized as such. But similarly important are the people around the
core innovators, people who are willing to play a supporting role in making
a great idea succeed.
While we all have the preconceived image of the genius in his experi-
mental cabinet cranking out inventions, the people surrounding the genius
are at least similarly important, forming the COIN working in swarm cre-
ativity to develop the innovation. Leonardo da Vinci had a whole support
organization of painters, sculptors, and other artists on his payroll that helped
to deliver his master works. Da Vinci treated and paid them well; in return,
they worked for his brand. Da Vinci would not have succeeded had he not
spoken in a language understood by his environment and had he not been
surrounded by a team of collaborative innovators. Of course, even da Vinci
swarm creativity
38
Sidebar 2.4
SwarmCreativity at Work in My First COIN
My experience with COINs began when I was a postdoctoral researcher at the
MIT Laboratory for Computer Science. I was part of a small group of people
that wanted to do something revolutionary for computer science education. Our
goal was to develop an educational multimedia CD-ROM that would convey
in a more interactive and visual way the highly complex subject of computer
algorithms---one of the hardest parts of computer science. We were not the first
researchers to produce computer animations for algorithms, but we were the first
seeking to build an integrated system.
We enjoyed two types of support. First, three MIT educators who had written
a popular book on algorithms gave us their full text, allowing us to base our
animations on their explanations and pictures.
1
Second, and more important,
one of these coauthors agreed to sponsor a class I was planning to teach
at MIT about algorithm animation and hypertext. We were able to build a
prototype system of animated algorithms as a series of class projects, and
that convinced our lab director to give us a small grant to turn the prototype into
a full commercial product. This enabled us to launch our little COIN, comprising
a highly motivated core team (working basically for free) and a team of a dozen
students. We used the grant to buy some hardware and to compensate students
for refining the algorithm animations produced in the course.
sometimes fell into the trap of being too much ahead and not communicating
in a language understood by his disciples. When he described parachutes,
planes, and helicopters, for instance, he was ahead of his time by hundreds
of years (gure 2.1).
We could say that da Vinci was the core of a self-organizing swarm of
similarly minded artists and scientists, working together in the creative chaos
of Renaissance Italy. The swarm used the same self-organizing properties
social insects employ: positive feedback, negative feedback, randomness, and
multiple interactions. It was a nurturing environment where collaboration was
amplied by positive feedback (for instance, da Vinci and Pacioli working
on their book on golden ratio, which neither could have completed on his
own). Early on, da Vinci honed his collaborative skills when he success-
fully dealt with negative feedback by other artists less brilliant than himself.
Collaborative Innovation through Swarm Creativity
39
The core team consisted of three people with almost no money but a great
deal of enthusiasm and excitement---that was enough to keep us motivated. The
students worked very hard, turning class project animations into production-level
computer visualizations stable enough for a commercial product.
After another year at MIT, I returned to Switzerland. Some of the students
moved to other universities. But while we were now geographically distributed,
the team remained on track. After two years of hard work our COIN published
a CD-ROM of Animated Algorithms to excellent notices from reviewers and users
alike.
2
This COIN experience showed that COINs are both powerful examples of
teamwork and great levers of individual strengths. The project would have been
impossible for any one of us working alone, but together our swarm creativity
was unbeatable. Our success was possible only because of the team spirit that
had developed in the COIN, where the whole team was driven to realize a shared
vision. The core team acted as creators, as collaborators who pulled together a
diverse team and kept it hard at work, and as communicators who convinced
sponsors and coworkers of the projects merits. And while we shared a common
set of characteristics, our individual strengths were complementary skills. We
worked hard not from a desire to make money, but from commitment to our
shared goal and a desire to succeed.
1. Cormen, Leiserson, and Rivest, Introduction to Algorithms (1990).
2. Gloor, Dynes, and Lee, Animated Algorithms (1993).
Similar to the positive use of chaos and randomness by insects, da Vinci
used the chaotic environment in northern Italy to maintain his independence,
to nd the most generous and venturous sponsors, and to gather around
him the best disciples. Finally, similar to ants that need multiple interac-
tions with each other to complete their tasks successfully, da Vinci could
excel only in an environment bustling with other painters, sculptors, math-
ematicians, and scientists building upon each others work in continuous
coopetition.
The diffusion of disruptive innovations generally follows the same pro-
cess. First, there is a genius with a brilliant idea. Working in relative isolation,
the genius develops the idea and implements the rst prototypes. But such
ideas are so far ahead of their time that the people around the genius do not
recognize their greatness. It takes a collaborative leader able to congregate a
swarm creativity
40
Figure 2.1. Da Vincis parachute design.
Copyright Biblioteca Ambrosiana, Milan.
Codex Atlanticus, f.1058v.
group of dedicated disciples to spread the message. This leader has a charis-
matic yet collaborative personality, is highly persuasive and imparts conta-
gious enthusiasm; he or she shepherds disciples toward the communal vision.
The motivated disciples infect the environment with their own enthusiasmand
keep things going until they reach the tipping point, where the new idea be-
comes part of daily life. Together, the collaborative leaders and the motivated
disciples constitute a COIN.
Familiar Examples of the COIN-Driven Innovation Process
There are many familiar examples of this COIN-driven innovation process,
even if readers may not realize the involvement of COINs. PCs, the Web,
the Internet, and Linux have already been mentioned. There were early
innovators such as Douglas Engelbart, who returned from World War II,
where he had been a radar technician, with a vision of working interactively
with computers. Engelbarts inventions were crucial for the success of both
the personal computer and the Web. Between 1963 and 1968, he invented
technologies to support individual knowledge workers and collaboration
within groups of knowledge workers. Out of his work came the computer
mouse, multiple windows on a screen, and e-mail. Another researcher, Ivan
Collaborative Innovation through Swarm Creativity
41
Sidebar 2.5
Torvalds Takes on the Critics of Open Source
Linux pioneer Linus Torvalds famously cited another great innovator in arguing
for the openness and collaboration intrinsic to COINs---the core method of open
source software development. Responding to statements by Craig Mundie, a
Microsoft senior vice president who criticized open source software for being
non-innovative and destructive to intellectual property, Torvalds sent out this
e-mail retort:
I wonder if Mundie has ever heard of Sir Isaac Newton? Hes not only famous
for having set the foundations for classical mechanics, but he is also famous
for how he acknowledged the achievement: If I have been able to see further,
it was only because I stood on the shoulders of giants. . . . Id rather listen to
Newton than to Mundie. He may have been dead for almost 300 years but
despite that he stinks up the room less.
1
1. See Wikipedia at http://en.wikipedia.org/wiki/Linus Torvalds.
Sutherland, created the rst graphics package, called Sketchpad, while
working at the Massachusetts Institute of Technology (MIT). Sutherland and
Engelbart were typical early genius inventors, dening a new eld with no
expectation of raking in nancial prots.
Collaborative leaders of the PC revolution include Alan Kay, Butler
Lampson, Robert Taylor, and Charles Thacker. They built the Altos, among
the rst personal computers, while at Xerox PARC. The rst of the motivated
disciples toclearlyrecognize the signicance of that inventionwas Steve Jobs,
whose Apple Computer Company popularized the concept of the personal
computer. Later disciples collected even larger rewards. People such as Robert
Noyce, cofounder of Fairchild Semiconductor and Intel, and Bill Gates,
cofounder of Microsoft, combine technical skills with uncanny foresight and
vision, the ability to create a highly cohesive company culture, and great
salesmanship. We can see how well this has repaid them!
The creation of the Internet, the global computer network that is the
foundation of the Web, delivers an even more compelling story of how a
group of intrinsically motivated people can drive open disruptive innovation.
swarm creativity
42
Figure 2.2. COIN-driven innovation process.
The Internets creation was very much driven by COINs. Soon after the
rst mainframe computers had been developed for military use in World
War II, a few visionary thinkers realized that, by linking these mainframes,
they could make much better use of computers. They were motivated not by
nancial reward or promotion, but by a deep desire to further technical devel-
opment. It was clear to those visionaries that the usefulness of the computer
network would grow exponentially with the number of linked computers.
The goal of the early Internet visionaries was to develop an open net-
working standard linking computers of any make and model anywhere on the
globe using publicly available networks such as the phone system. J. C. R.
Licklider, or Lick, as friends, colleagues, and acquaintances called him,
was the best known among the pioneers; he had a vision of allowing multiple
users to simultaneously use the expensive mainframe computers of his time.
Newly appointed in 1962 to head the Computer Research Division at the
U.S. Defense Departments Advanced Research Projects Agency (ARPA),
Collaborative Innovation through Swarm Creativity
43
Figure 2.3. COIN leaders for the evolution of personal computers.
Licklider oversaw the biggest computer-related research budget of the U.S.
government. Against the advice of the computer establishment, he used his
position to mastermind time-sharing computers and then to fund develop-
ment on the Advanced Research Projects Agency Network (ARPANET), thus
single-handedly creating internetworking. In 1960, he accurately foresaw the
Internet, envisaging thinking centers, that will incorporate the functions of
present-day libraries together with anticipated advances in information stor-
age and retrieval. . . . The picture readily enlarges itself into a network of
such centers, connected to one another by wide-band communication lines
and to individual users by leased-wire service.
10
Contemporaries describe
Licklider as a man who retained his modesty despite his considerable in-
uence on computing. The list of people he selected and put in charge of
implementing his vision at research universities that included MIT, Carnegie
Mellon, the University of California at Berkeley, and Stanford reads like a
Whos Who of todays computer research leaders.
The people who further extended the ARPANET into the Internet formed
a COIN in its purest form. A self-organizing network of individuals, the
swarm creativity
44
Sidebar 2.6
The Dedicated Disciples Who Carry Innovation over the Tipping Point
In his best-selling book The Tipping Point, Malcolm Gladwell describes the
characteristics of the dedicated disciples it takes to make a great idea a great
success.
1
These connectors, mavens, and salespeople are a crucial part of the process.
The connector knows many different people in different socioeconomic areas.
For example, Paul Sachs, an heir to the Goldman Sachs banking empire, was a
connector. When the Museum of Modern Art (MOMA) was founded in New
York in 1929, Sachs played a crucial role, spanning many of what social scientist
Ronald Burt calls structural holes to connect the modern arts world and New
York high societys wealthy benefactors. Later, Sachs became an art professor
at Harvard University and recommended one of his best students, Alfred Barr,
for the job of first MOMA director. He also assisted the museums founders in
finding potential sponsors, board members, and works of art. Paul Sachs was
an archetypical connector, linking the distinct fields of high society, finance, and
modern art.
Mavens are quite different from connectors. They do not necessarily have
to know many people, but they are experts in a certain field and---driven by
altruism---they love to provide this expertise to others. Mavens are helpful
connoisseurs who carry the message of a great innovation because they are
convinced it is great. The very nature of this supportive task makes mavens much
less visible (usually) from the outside. We encounter small mavens when we get a
tip about the best restaurant in a certain neighborhood, or when someone helps
us connect complicated technical equipment.
We have all met the salespeople who comprise Gladwells final category. They
can sell just about anything---not through pressure, but because they have
emotional intelligence.
2
Real salespeople feel what their clients want and
are able to package and present their products in a way that profoundly appeals
to those clients. Great politicians are usually excellent salespeople; both Ronald
Reagan and Bill Clinton had an uncanny sense for the concerns of the common
man. They stayed in close touch with the electorate while in the White House.
This had a lot to do with why they won second terms and enjoyed unprecedented
job approval ratings.
1. Gladwell, The Tipping Point: How Little Things Can Make a Big Difference (2002).
2. Goleman, Emotional Intelligence: Why It Can Matter More than IQ (1997).
Collaborative Innovation through Swarm Creativity
45
Figure 2.4. COIN leaders for the creation of the Internet.
Internet Engineering Task Force (IETF) and its predecessor organizations
have steered the development of the Internet for the past 30 yearswithout
any central authority. Anybody could (and still can) join an IETF working
group.
The Internet Engineering Task Force is a loosely self-organized group
of people who make technical and other contributions to the engineer-
ing and evolution of the Internet and its technologies. . . . There is no
membership in the IETF. Anyone may register for and attend any meet-
ing. The closest thing there is to being an IETF member is being on the
IETF or working group mailing list.
11
There have been hundreds of IETF working groups since its rst meeting on
January 1986 in San Diego, California, with 15 attendees. The predecessor
of IETF, the Network Working Group (NWG), goes back 18 years earlier.
Those working groups have been developing seminal standards and public
domain implementations like transmission control protocol/Internet protocol
swarm creativity
46
(TCP/IP), the main communication protocol of the Internet and Internet mail,
and the protocol that manages millions of hosts on the network. Each IETF
working group functions as a COIN, where a small core teamdriven by
visionary leaders such as Vinton Cerf and David Clarkcollaborates on de-
veloping the specications. Meanwhile, other members act as a sounding
board, reading and commenting on the specs and testing the so-called refer-
ence software implementations developed by the core team. IETF working
group participants receive no nancial compensation. They are normally on
loan from a company research lab, are paid by an academic institution, or are
self-employed consultants. Their main motivation to participate is the desire
to be part of the further development of the Internet and to advance the state
of the art in internetworking technology.
These individuals share this intrinsic motivation with the group of pro-
grammers that created the World Wide Web, which evolved in a similar way
to the Internet. Early geniuses such as Vannevar Bush, Ted Nelson, and
Douglas Engelbart laid the groundwork by inventing technologies needed
for the Webs success, such as linking, hypertext, multiple windows, and the
computer mouse. Collaborative leaders such as Tim Berners-Lee and Robert
Cailliau recognized the importance of these inventions and put the pieces
together. But, again, it took motivated disciples to carry the innovation over
the tipping point and trigger the tidal wave of e-business. Netscapes Marc
Andreessen, e-Bays Pierre Omidya, and Amazons Jeff Bezos all combine
the technical prowess of a maven with the wide people network of a connec-
tor and with great salesmanship. Not only did they launch highly successful
companies, but they also redened their respective industries and reshaped
their competitive landscapes in the process. They share these characteristics
with the open source programmers who created Linux.
Collaboration between Innovators Is the Key
Fritz Bircher and Reinhold Krause in Switzerland are perpetual innovators.
That makes themperspicacious COINmembers. But besides being very smart
and having creative ideas, they are also excellent team players. This means
that, in their COIN that formed around GPS-controlled lawn mowers and
MobileJet printers, they have built up a collaborative team culture, based on
Collaborative Innovation through Swarm Creativity
47
Figure 2.5. COIN leaders for the creation of the Web.
meritocratic principles. This does not mean that hierarchy is absent; to the
contrary, it is clear that Bircher and Krause are the leaders. But they give
credit where credit is due, and they foster a transparent work environment
with high respect for the individual.
For the success of a COIN, it is the middle phase of the invent-create-sell
process that is most critical. The MobileJet would never have been possible
if Bircher and Krause had not collaborated creatively. In this second, collab-
orative creation phase, Krauses mechanical engineering skills and Birchers
software and electrical engineering skills were pooled together to carry a truly
innovative product from concept to implementation, transforming it from a
lofty idea into a working solution. The interaction between Krause, Bircher,
and the 12 graduate students and research assistants in their COINat Burgdorf
was the catalyst inthis crucial phase. Once theyhadsomethingtodemonstrate,
Bircher and Krause communicated with the outside world. Their innovations
have been featured on Swiss television, for instance, which has in turn led to
contacts from potential users and investors.
swarm creativity
48
For a COIN to succeed and get its ideas off the ground, there must be
creativity, collaboration, and communication. Only a combination of people
who bring these traitsthe DNA of COINsto the team can succeed in
bringing a COINs innovation over the tipping point. That DNAis the subject
of the next chapter.
CHAPTER3
The DNA of COINs
Creativity, Collaboration, and Communication
About 30,000 years ago, something amazing happened in Europe. For the pre-
vious 15,000 years, modern Homo sapiens, our ancestors, and Neanderthals
had been living together. But then, in a prehistoric competition for survival,
the Neanderthals disappeared. Homo sapiens, the survivors, went on to expe-
rience a cultural revolution, making great progress in art and craftsmanship.
Why did this happen? While Neanderthals were living in the same caves
for generations, roving bands of our ancestors followed the herds, traveling
great distances. During their travels, they met other people, exchanging ideas
and learning about newinnovations. Harsh external living conditions brought
by the last Ice Age in Europe forced Homo sapiens, and those they met,
to cooperate and innovate. While this external stimulus led to a urry of
innovations by Homo sapiens, it also led to the extinction of the Neanderthals.
So Homo sapiens beat out Neanderthals because of learning networks
this is the rst of the ve key elements of the genetic code of collaborative
innovation networks (COINs).
Over the next 20,000 years, modern Homo sapiens progressed tremen-
dously. The prehistoric trading camps grewinto large cities housing thousands
of people all around the globe, fromThebes in Egypt, to Ur in Mesopotamia, to
Changan in China, to Rome in Italy. Beginning in Roman times, the Chinese
49
swarm creativity
50
and Roman empires were connected by the Silk Road, trading goods and
knowledge for over a thousand years. But with the demise of the Mongolian
and Byzantine empires toward the end of the European Middle Ages, trafc on
the SilkRoadcame toanend. One of the last travelers of the SilkRoadwas also
one of the most famous: Marco Polo carried back to Europe knowledge about
the Mongolian Imperial Chinese court, and about Chinese culture and daily
life such as the Imperial postal system and paper money. And while the Silk
Road was closed down because local warlords were ghting with each other,
Renaissance Europe experienced a boom in science, culture, and commerce.
Almost a contemporaryof MarcoPolo, GermanRenaissance banker Jakob
Fugger rose to become the richest man of his time. At the height of its wealth,
the Fugger banking empire owned a tenth of all money owing in the Holy
Roman Empire, which stretched fromtodays Poland, Germany, Italy, Nether-
lands, Spain, and Portugal to Latin America and the Southeast Asian spice
islands. While Emperor Charles V rightly claimed that in his empire the sun
never set, he did not openly proclaim that it was Fugger who put him on the
throne.
Jakob Fugger was not just an uncommonly shrewd businessman. He was
also a great benefactor to the poor and needy and took excellent care of his own
workers. He founded hospitals and elderly homes for his own employees, and
he gave freely to the church and to the poor. In his hometown of Augsburg, he
constructed the rst social housing project, a subsidized apartment village for
needy families who had to pay only a nominal rent. The Fuggerei still exists
today, and it is still nanced by the Fugger family; likewise, the Fugger private
bank still operates, and the Fuggers are still a very wealthy family. Other Euro-
pean bankers and barons who competed with Fuggers bank for the favor and
the purse of the emperor are gone; they cared only about their own well-being;
and rarely did their fortunes signicantly outlast the founding generation.
Sustainable business success requires a combination of keen business
sense with sound ethical principles. That is the second key element of the
genetic code of COINs. The third is trust and self-organization, the very
things upon which Renaissance Europe thrived.
In 1449, Jakob Fugger and another rich merchant from Augsburg named
Bartholomus Welser nanced the prospecting expeditions of Pedro Alvarez
Cabral to Brazil, enlarging their immense fortunes in the process. At the same
time, the Chinese emperor burned his huge eet and decided not to venture
beyond his mainland empire. And the diffusion of the invention of gunpowder
The DNA of COINs
51
provides a fascinating example of howxenophobic isolation can be disastrous.
The Arabs and Chinese had invented gunpowder centuries before medieval
scientist and Franciscan friar Roger Bacon rst documented the formula in
Europe, but the Chinese were not able to put gunpowder into productive use.
They used it for reworks, and even built crude gunpowder weapons, but the
development of rearms in China stagnated until the Chinese encountered
them again in a stunning defeat by the invading Europeans in the 19th cen-
tury. Meanwhile, the warring city-states of tiny Renaissance Italy provided
fertile grounds for nurturing the invention of the rearm. Charismatic Italian
Renaissance potentates like Lorenzo di Medici or Pope Alexander VI Borgia,
who relentlessly fought each other for political supremacy, provided huge
incentives for creative innovators. The simple explanation for why China,
a hundred times more powerful, stagnated in developing this sophisticated
technology is that Chinese isolationist policy did not provide for the pro-
voking, stimulating and at times chaotic hotbed environment that European
Renaissance inventors such as Leonardo da Vinci faced. What the Ice Age
did to Neanderthals, isolationalist policies did to imperial China. Satised
with the status quo, the Chinese were overtaken by more nimble, innovative,
creative, and chaotic competitors. China was ruled by an authoritarian em-
peror and a strictly regulated hierarchy of Confucian magistrates, while Italy
and Germany were ruled by creative chaos and self-organization. People
swarmed around petty rulers, bankers, and merchants acting as patrons of
artists and innovators who autonomously took hold of their own destiny.
In Europe, a system of medieval guilds and cities took care of their own
guild members and citizens; they created a system based on self-organization
and trust. Nothing of the sort existed in China.
1
In Europe, craftsman guilds,
merchant networks, courts of the gentry, and town communities functioned as
interpersonal trust networks, providing a exible foundation to develop new
inventions at low costs for society. To survive, innovators had to collaborate
and share what they knew within their trust networks.
The fourth key element in the genetic code of COINs, making knowledge
accessible to everyone, is well illustrated by another European example. By
the fteenthcentury, the Catholic Churchandthe Popetrustedspiritual lead-
ers and the heart of the barbarian Western countries for over a millennium
had become a rampant hierarchical ulcer. Trust was perverted by an obscure
and complicated hierarchy, individual greed, and corruption. Along came
Martin Luther, who started his career as a simple priest in Germany, far
swarm creativity
52
from Rome. His Reformation began with questioning of the idea that timely
salvation could be brokered by the church in the form of indulgences. Such
egalitarian, anti-authoritarian habits of thought led him to more general as-
sertions, nally giving ordinary people direct access to the word of God
without the need of interpretation by priests. Luther sought to shift the power
of spiritual authority back to where he felt it belonged: the hands of the
ock. His ideas spread like wildre. Today, Luthers protestant church has
split up into smaller entities, becoming more transparent and operating by
swarm intelligencewithout a single, pope-like leader, where community
members contribute their time and skills based on individual motivation in
self-organizing knowledge networks.
The success of the Reformation was driven to a large extent by Johannes
Gutenbergs invention of metal type. In 1455, Gutenberg published the Guten-
berg Bible in Mainz, Germany. For the rst time, the Bible became affordable
not only to the rich, but also to the middle class. There was still an obstacle,
thoughit was printed in Latin, and the ordinary layman needed the clergy
to translate the Word of God. Martin Luther changed this when he translated
the Bible from Latin into German; his translation was rst printed at Wit-
tenberg by Hans Lufft in 1534. It is said that Lufft sold over 100,000 copies
of Luthers Bible in 40 years. Thanks to the invention of Gutenberg, and the
egalitarian spirit of Martin Luther, religious communication ourished and
crucial knowledge was made available and placed back into the hands of the
community. Knowledge had been made accessible to everyone.
The fth key element of COINDNAis internal honesty and transparency.
This is illustrated by the story of the Rothschilds.
Collaboration, knowledge sharing, trust, and reciprocal principles have
been a recipe for sustainable successnot only in religion, but in business
dealings back to the times when business was conducted by bartering. Forbid-
den to own land or join craft guilds, medieval Jews were conned to banking
and nance. They had created a pan-European banking system based on re-
ciprocal trust, in which Jewish bankers were able to draw on accounts of
fellow Jews in foreign cities, knowing they would have to reciprocate the
favor. This system worked perfectly, based on a mutually established code
of ethics that forbade cheating. It reached its climax when, more than 200
years after Fugger, the Jewish banking dynasty of the Rothschilds created a
banking empire that similarly dominated the European nancial system for
almost a century.
The DNA of COINs
53
Born in 1744 in Frankfurt, Germany, Mayer Amschel Rothschild sent his
ve sons to the major European cities. Relying on trusted family members
in all the European nancial centers, the Rothschild bankers not only were
able to transfer money quickly without having to transport it physically,
but they also constructed an efcient information network based on carrier
pigeons linking their different branches in major European cities. The second
generation of the Rothschilds accumulated immense wealth in London, Paris,
Vienna, Frankfurt, and Naples.
The Rothschilds followed their own ethical code of internal transparency
and reciprocity. Each family member had to informthe others openly about his
business dealings, while keepingnon-Rothschilds out (onlythe male members
of the Rothschild family were let into the innermost banking circle). For
more than 100 years, the close-knit Rothschild banking empire dominated
European banking, until the telegraph democratized access to information
for everybody.
Over the course of history, these ve elements have characterized teams
of collaborative innovators, starting at the dawn of humanity. Disruptive in-
novations like gunpowder, the printing press, paper money, and bookkeeping
are not the result of a carefully planned innovation process. Nevertheless,
by analyzing teams of collaborative innovators of the past, we can iden-
tify their typical behavioral patterns. Those lessons from the past form ve
corollaries.
1. Collaboration networks are learning networks.
2. Collaboration networks need an ethical code.
3. Collaboration networks are based on trust and self-organization.
4. Collaboration networks make knowledge accessible to everybody.
5. Collaboration networks operate in internal honesty and transparency.
The elements constitute the genetic code of participants in COINs. And
this genetic code of creativity, collaboration, and communication is deeply
ingrained in the members of the cyberteam that make up a COIN. Linux
developers offer us a modern example (see sidebar 3.1).
While the concept of innovation teams ranges back in history, the con-
nective power of the Internet has greatly sped up the dissemination of new
ideas, giving charismatic leaders with innovative schemes immediate global
reach. But it is still the brilliant genius who comes up with the new idea, and
swarm creativity
54
Sidebar 3.1
The Genetic Code of the Linux Debian COIN
Debian is the nonprofit distribution of Linux, competing with commercial
distributions sold by Linux vendors such as Red Hat or SuSE. The Debian
software developers are a vibrant community that has reinvented itself at
least six times during the last eight years---exchanging leaders, changing how
leaders are elected, administering its own social contract, creating the Debian
constitution, and overhauling the voting procedure. In the process, it has become
an increasingly efficient community.
The Debian community makes sure that its members share the right genes as
highly skilled developers. Applicants first need to prove their worth by recom-
mending changes. If those changes are accepted, a subsequent application must
be sponsored by an accredited developer. Officially initiated members of the De-
bian COIN demonstrate five collaborative genes in an exemplary fashion. Motiva-
tional drivers like those introduced in the previous chapter---to satisfy ones ego,
gain more knowledge, and have fun---blend seamlessly into five characteristics:
1. There is a willingness to experiment.
2. The Debian social contract is an explicit code of ethics.
3. COIN members follow the Golden Rule, which builds mutual trust.
4. An essential credo is to make knowledge accessible to everybody.
5. The entire COIN is based on internal honesty and transparency.
In fact, the Debian community goes much further than, say, the Rothschilds,
committing itself not only to internal but also to public transparency. For
instance, the Debian social contract explicitly states We Wont Hide Problems,
and all problems are published on a public Web site.
the group of collaborative leaders who get their team of disciples together to
spread out the wordthe Internet is only the catalyst.
Collaborators in Todays Knowledge Networks
Homo sapiens, Fugger, Renaissance Europe, the Reformation, and the Roth-
schilds exhibit characteristics of collaborative innovators working together
in knowledge networks well before there was an Internet. They all worked
The DNA of COINs
55
as teams that embraced change, performing together in a collaborative way
while conducting their business according to the ethical principles of their
time. They shared knowledge internally, always ready to learn and to adapt to
a changing environment. They share these traits with modern-day collabora-
tive innovators such as open source programmers, the group that developed
the World Wide Web, or the Internet Engineering Task Force (IETF) Internet
standardization group.
There are some basic differences between the pre-Internet innovation net-
works and the Internet-enabled COINs of today. Members of innovation net-
works of the pre-Internet age had a fundamentally different attitude toward
authority, transparency, and meritocracy. But the biggest difference is the ex-
istence of the Internet, which gives todays community members immediate
global reach at very low cost; it allows teams to collaborate at a level of
transparency never before possible. Unlike in earlier history, todays innova-
tions do not trickle down through communication by carrier pigeons. COINs
have reached their own tipping point, disseminating new ideas and concepts
on a global scalenot in centuries, but in days and weeks. Linux and the
World Wide Web became success stories within just a few years. Modern-
day communication capabilities act as a huge catalyst, with the Internet giving
the nal boost. As instantaneous communication with anybody to anywhere is
essentially free over the Internet, trust building, collaboration, and exchange
of ideas can all happen at the speed of light. Asecond fundamental difference
is that, whereas leaders of early innovation networks demanded unquestioning
authority, the Internet-enabled COINs of today operate as true meritocracies.
While Tim Berners-Lee and Linus Torvalds have reached guru status, mem-
bers of the Web consortium and Linux developer communities obtain higher
status through outstanding contributions and hard work transparent to all. In
the Fugger and Rothschild banking empires, promotions were based on the
decisions of family patriarchs.
The people making up todays COINs normally cross organizational
boundaries. They form a dedicated community of members existing outside
conventional organizations but sharing a common vision. Membership and
roles within the community are based on contribution and merit, not on exter-
nal hierarchical status. The community develops its own internal hierarchy
based on skills and contributions, and people join the community not for
immediate monetary reward, but because they believe in the goals of the
community. Todays COIN members follow their leaders not because those
swarm creativity
56
leaders demand unconditional hierarchical authority, but because of shared
beliefs, values, and goals. Thanks to the speed and open architecture of the
Internet, todays innovation networks operate in internal transparency, and
leaders are continuously challenged by their community. In the 21st century,
the roles of COIN leaders and members are converging.
The essence of a community is its members. Membership in COINs is
voluntary, rather than prescribed. Members are self-organizing and participate
because they get high personal satisfaction from their participation; they
are highly self-motivated and committed to their cause. Additionally, they
are intellectually curious, are opposed to opaque hierarchies, and have a
strong egalitarian mindset. Intels Lablet COIN is a superb example (see
sidebar 3.2).
But what about the communication aspects of COINs? What are the
mechanisms by which people in COINs connect with others in their COIN
and with the outside world? How are innovations disseminated?
Marco Polo: Collaborative Communicator of Innovations
Marco Polo may be the most famous world traveler of all time. His travel
reportswritten in a factual style that set him apart from his contempo-
raries
2
became part of the mental climate of Europe. His writings shaped
the perception of European geographers, adventurers, humanists, and Jesuits
for centuries, as they set out to discover the world. The learned fromall over
Europe consulted him in Venice on aspects of academic geography. During
his lifetime alone, Polos book was translated into French, Franco-Italian,
Tuscan, Venetian, Latin, and Germanunprecedented for a book published
in the Middle Ages.
Polos travel reports stuck to the facts. He kept himself in the background,
and he focused not on adventure but on providing a handbook for travelers
and merchantsfor example, giving the rst detailed account to the West on
the provenance of spices and other trade goods. This approach was immensely
effective. As a collaborative communicator of innovations, he succeeded in
changing the world. As an archetypical communicator of innovation, Polo
followed four simple rules and intuitively established four principles of open
and honest communication:
The DNA of COINs
57
Sidebar 3.2
Innovation Networks at Intel Research
Intel Advanced Research consists of a collection of COINs that communicate
with each other in a small-world network. Under the leadership of David
Tennenhouse, the companys director of research, Intel outsources a crucial
part of its advanced research to its network of Lablets---small research labs,
co-located at prominent universities such as the University of California at
Berkeley, the University of Washington---Seattle, Carnegie Mellon University, and
the University of Cambridge in England. Lablet staffs comprise an equal share of
(a) professors and doctoral students on temporary leave from the university and
(b) Intel employees. Although Tennenhouse and the Lablet directors define the
overall strategic research agenda, researchers at the Lablets have wide-ranging
freedom to pursue their own interests. Research teams, usually spanning different
Lablets, form their own smaller COINs, while the combined Lablets form a
knowledge network that reaches into the hosting universities and the central
Intel research labs.
PlanetLab is one of Intel Advanced Researchs early successes. A network of
computers located at sites around the world, PlanetLab forms a platform for
creating and deploying planetary-scale services---massive applications that span a
significant part of the globe---such as global content distribution and file sharing.
Applications running on PlanetLab are decentralized, with pieces running on
many machines spread across the global Internet. They can also self-organize
to form their own networks. No university could afford to deploy the large
network of computers required to create PlanetLab, so Intel jump-started the
project by contributing the first 100 nodes, the initial software implementation,
and key sites. After some years of nurturing in the Intel Lablets, digital video
content distribution has been announced by Intel and Hewlett-Packard as a first
PlanetLab commercial application.
Intel Lablets leverage the latest results of high-tech research. Teams collab-
orate to achieve shared goals, while concurrently competing with and learning
from their competitors. The rules are transparent and clear for everybody, and
the teams function as a meritocracy.
swarm creativity
58
Make the message objective. Unlike some other travel writers of his
time, Polo objectively portrayed the lifestyles of those with whom
he came in contact. He held no grudge against the Mongols, for
instance. His descriptions included truly useful tips for his fellow
merchants and voyagers.
Walk the talk. Marco Polo really did spend 24 years with the
Mongols. His 17 years at the court of the Great Khan gave him
unique insights into how the Mongolian empire was ruled, which he
shared with his readers.
insna/INSNA/socnet.html.
The Tao of COINs coordinates the working behavior of COIN afliates.
Usually COIN members carry this code in their genesthat is, they live by
it without it being written out. Behaving according to these mostly unwritten
ethical rules, COIN members are implicitly more sociable than members
of many societies that live by a written rigorous law. As COIN members
participate of their own free will and usually are not paid to work, they expect
and exhibit fairness, compassion, and altruism. COIN members go out and
help others because they expect others to do the same for them.
As the term implies, the Tao of COINs subsumes altruistic behavior,
compassion for fellow COIN members, and moderation and humility of their
Ethical Codes in Small Worlds
79
leaders. Three fundamental terms of a culture for COINs summarizes the Tao
of COINsmeritocracy, transparency, and consistency.
Meritocracy as an extension of reciprocity, the principle of fair treatment
of everyone, is at the core of successful innovation communities. COIN
members are willing to help and share what they know with others, but
they expect similar behavior from other COIN members. Meritocracy is a
direct consequence of the application of the Golden Rule. For example, open
source software developers contribute their code because they expect their
codevelopers to do the same so that all may enjoy the benets of using
a common software code base. The community judges the quality of each
individual contribution, since each contribution is open to the entire COIN.
Transparency means that rules are made explicit, and the roles and re-
sponsibilities of every COIN member are obvious to the entire community.
In a COIN, strengths and weaknesses of every member are exposed, but
contributions are also made transparent. For example, the skills and the role
of every programmer are obvious to all members working on a common open
source software project. Each team member expects to get fair credit for her
or his contributions; peer recognition is one of the main motivations for COIN
participants.
Consistency means that every COIN member behaves according to the
Tao of COINs, and delivers on promises made to the community. Under the
same conditions, similar actions will produce the same results. Consistency
is the basic requisite for fair treatment of all COIN members. For example,
open source developers are expected to stick to the programming rules and
guidelines that are in effect for their project. Members of the Debian Linux
distribution have dened the Debian Social Contract to ensure consistent
behavior of all COIN members, including their leaders.
Meritocracy, transparency, and consistency also mean that actions within
the community are grounded in reason and not in randomness. Innovation
communities are driven by learning, logic, and a shared vision of working
toward furthering the state of the art. In an open source software project, it
is expected roles will be lled strictly on the basis of merit, and not because
of previous relationships, hierarchical positions, or other criteria not relevant
to the project. COIN members behave along those guidelines without being
aware of doing so. The Tao of COINsdened by meritocracy, transparency,
and consistencyis part of their genetic code in the same way social insects
exhibit swarming behavior, enabling them to work toward a larger goal as a
swarm creativity
80
Sidebar 4.2
You Might Think Theyre COINs, but . . .
Lets look at organizations that are almost COINs. They may have the innovators
gene and operate with swarm intelligence, and they may have built virtual
communities that rely on the Internet for internal communication. But without
meritocracy, consistency, and transparency---essential for success---they arent
really COINs.
One such organization is the U.S. National Aeronautics and Space Adminis-
tration (NASA), created to innovate. But its innovative capabilities are severely
hampered by its rigid command and control structures. Open and honest
communication is buried in NASAs huge hierarchy. The New York Times posits
that the Columbia space shuttle disaster would not have happened if NASA had a
more collaborative, meritocratic, and transparent culture.
1
In a strong indictment of NASA management, crash investigators reported
that the agencys work culture discouraged dissenting views on safety issues.
The missing flow of information and the agencys approach to peer review and
accountability prohibited its own engineers from following their instincts. For
some of the deep technical analysis needed to make space flight successful,
NASA was relying on external contractors who were afraid to speak out in clear
language (an effect these contractors call NASA chicken syndrome). These are
characteristics totally adverse to COINs.
The worldwide antiglobalization movement offers another illustration. The
nongovernmental organizations (NGOs) and affinity groups that protest
self-organizing team, where each team member does not have to know the
big picture in detail all the time.
Occasionally, organizations may appear to be COINs, or like COINs,
because they display some of the typical characteristics. But the deciencies
can make all the difference, as sidebar 4.2 explains. When organizations
lack meritocracy, consistency, and transparency, it is an indication of another
serious deciencythat of social capital, which is for society what nancial
capital is for the economy. Investment in social capitalreciprocity in social
networksis the key building block for a COINculture. And an organization
founded on meritocracy, consistency, and transparency accumulates social
capital.
Ethical Codes in Small Worlds
81
globalization, organizing demonstrations with huge turnout at events such as
the G8 summits, operate much like COINs. Participants are members of a huge
shared-interests network, unified by their disdain for our current industrial
system of a globally linked economy, and in their rejection of the existing
system of world governance. A smaller group of more active practitioners
who organize the protests form a hard core. The movements nucleus forms
a COIN-like structure, with people sharing a common vision and cooperating
actively toward their shared goal.
Antiglobalization protesters have much similarity to COINs. They run a
creative organization, making highly efficient use of the Web to coordinate on
a global scale with high agility. But they operate shrouded in secrecy, lacking
the transparency that is a crucial success criterion for high-functioning COINs.
This lack of transparency also masks that the movement is short on alternatives
to the existing world order. The laudable vision of the movement to break down
existing wealth inequalities is not coupled with many specific goals on how to
bridge the divide---which means that the goals are not known and shared, as in a
COIN. If the movement wanted to operate successfully as a COIN, it would have
to develop more convincing goals, while at the same time laying open its ethical
code, governance and leadership structure.
Finally, there are religious cults, which have strong similarities to COINs. They
share self-organization and swarm intelligence, and they have the egalitarian
properties of communes. But they lack the rationality and transparency essential
to high-functioning COINs.
1. Leary, Better Communication Is NASAs Next Frontier (2004).
Social Capital Is the Currency of COINs
I have a friend who is a craftsman and who also breeds and races horses.
Horse racing is a very expensive hobby. My friend is not particularly rich
in nancial capital, but he is rich in social capital: he is embedded in a rich
web of social relationships and of trust, of giving and taking. He uses his
skills as a craftsman and horse owner to do favors and build relationships
with othersfor example, by helping xing little things in the household, or
by allowing children of friends to visit his horse stable. In return, his horse
trailer is serviced for free, and he gets materials for rebuilding a stable at a
very low price. He is not trading favors with people on a one-to-one level,
swarm creativity
82
but he is helping others in the expectation that others in his peer group will
reciprocate those favors. His wealth of social capital more than offsets his
nancial limitations.
Social capital can also be looked at from a group perspective. The social
capital of a group refers to the collective value of all social networks and the
inclinations that arise from these networks to do things for each other. Social
capital is the application of reciprocity in social networks, where individuals
are acquiring social capital by investing time, labor, knowledge, and other
nonmaterial goods in their community in the expectation that the community
will pay them back in kind.
Harvard sociologist Robert Putnam denes social capital as including as-
pects of social organizations, such as trust, norms, and networks that can im-
prove the efciency of the organization by facilitating coordination.
8
Another
way to dene social capital is to look at it as a stock of emotional attachment of
individuals to a group. Social capital can be positive or negative. Horizontal
networks of individual citizens and groups that enhance community pro-
ductivity and cohesion are positive social capital assets, whereas self-serving
exclusive gangs and hierarchical patronage systems that operate against com-
munitarian interests can be thought of as negative social capital burdens on
society.
Meritocracy, Consistency, and Transparency Build Social Capital
There are different ways to regenerate social capital that can be instituted at
the organizational level but target the individual. In the context of virtual
communities, the rst step toward building social capital is to improve
the education about, and awareness of, virtual teamworking in cyberteams.
Knowledge of Internet etiquette and practice in Internet technologies are
prerequisites for effective participation in virtual teams. Improved awareness,
access, and education on virtual teamworking will lead to a more participatory
workforce. A second step is to encourage increased participation in activities
outside of the core business and to back community-oriented workplace
practices. Deconstructing huge business units and creating small and exible
teams comprising a we-feeling supports this more active involvement
of individuals into non-core activities. Supporting the creation of social
capital also means that rms should act responsibly toward their employees
Ethical Codes in Small Worlds
83
Sidebar 4.3
ABroader Viewof Social Capital
The World Bank takes a very broad view on social capital, in which social
capital refers to the institutions, relationships, and norms that shape the quality
and quantity of social interactions.
1
In this view, social capital is not only the
sum of the institutions that underpin society---it is the very glue that holds them
together. Social capital is a companys stock of human connections, such as
trust, personal networks, and a sense of community; it is so integral to business
life that, without it, cooperative action and productive work are impossible.
Almost every managerial decision---from hiring, firing, and promotion, to
implementing new technologies, to designing office space---is an opportunity for
accumulating or losing social capital. To grow social capital in an organization
means hiring and encouraging people who fit the values and culture of the
organization, and creating an environment in which social capital will build.
To do this, companies must take active steps to develop trust, networks, and
communication. The benefits of this approach will be better knowledge sharing,
lower transaction costs, lower turnover of key employees, better coherence of
action due to organizational stability, and more shared understanding. There will
also be more creativity if people are allowed to experience the intrinsic pleasures
of taking the future into their own hands. Commitment and cooperation will
increase, and customer needs will be more intelligently answered.
1. See http://lnweb18.worldbank.org/ESSD/sdvext.nsf/09ByDocName
/SocialCapitalInitiativeWorkingPaperSeries.
families and community commitments and encourage other employers to
follow their example. For instance, accommodating part-time work increases
social capital, since part-time work increases ones exposure to wider social
networks while leaving enough time to pursue other interests outside the
workplace.
Wealth in social capital in an organization benets both the individual and
the group. This means that collaborative actionandcoordinationis made much
easier in an organization where high social capital has been accumulated.
An organization whose members have accrued high levels of social capital
provides a fertile nurturing ground for COINs, because these organizations
exhibit strongly three characteristics linked to the accumulation of social
capital:
swarm creativity
84
The period when the new innovation team was most active was easy
to identify, and it was also easy to identify the core members of the
new innovation team.
This analysis was done two years after the data had been collected. Had
the tool been available in real time to monitor virtual interaction, senior
management would have been afforded the opportunity to provide adequate
support for the new COIN. That, in turn, would have reduced time to market
for the innovation and increased awareness of the teams output within the
consulting rm. In short, not being aware at the time of what was going on
likely cost Deloitte consulting revenue.
Lets consider some specic examples of howTeCFlowanalyzes different
patterns of knowledge in different situations.
Innovation Pattern: Creation of a New Service Offering
The following example illustrates the interaction pattern of an innovation
community. We are analyzing a subset of the Deloitte e.Xpert messages used
for the analysis in the previous section of this appendix. The subset was
appendix b
155
Figure B.12. Visualization of an entire e.Xpert practice.
categorized manually to include all communication about the creation of the
new CKN diagnostic service offering, which resulted from the innovative
work of the COIN identied above.
A dedicated team of 39 volunteers from eight countries around the globe
worked together to create this new consulting service offering. The core team
appendix b
156
members came fromSwitzerland, the UnitedKingdom, the UnitedStates, Bel-
gium, and Germany. They never met face-to-face, but they worked together
over a six-month period to create a software platform and marketing material
that were widelyusedtowinconsultingengagements over the next three years.
The rst TeCFlow movie snapshot in the upper left of gure B.13 illus-
trates the low GBC and high-density structure of the innovation team (or
COIN). The trough in the GBC plot in the lower left of gure B.13 shows the
initial phase where the teamcollaborated to create the offering, while the peak
toward the end of the GBC plot demonstrates the structure with a tightly con-
nected core surrounded by weakly connected people (i.e., the innovation team
was encircled by a learning and interest community). This communication
pattern is shown in the movie snapshot at the upper right of gure B.13.
The contribution index plot in the lower right of gure B.13 demonstrates
the immense efforts of the practice coordinator, who was vital for the projects
success; this contribution was not apparent at the time the service offering
was created. There is a huge activity gap between the practice coordinator and
the rest of the project team. A summer intern hired to help create this offering
was only peripherally integrated into the core team, as the analysis shows,
because team members did not adequately include her in the communication
ow of the project.
Learning Pattern: Information Dissemination by Webinar
The next example illustrates how new concepts are taught to an online
audience forming a CLN. Our data set for this example consists of an e-mail
archive of the e.Xpert consulting practice, where a team was organizing a
global Web-based seminar (Webinar). The archive covers a period of about
200 days.
Asmall teamof self-selected members of the e.Xpert practice prepared the
Webinar over several months. One main speaker then delivered the Webinar
over one hour, assisted by his team members. The audience, spread out
globally, had the opportunity to question the speakers via e-mail during the
Webinar. Questions that could not be answered during the talk were replied
to over the next few days, and because of overwhelming demand, the team
decided to rerun the Webinar a few weeks later after making some minor
changes. The rerun had a different main speaker.
appendix b
157
Figure B.13. Expansion of a COIN into a learning and interest community.
The TeCFlow movie illustrates nicely the switch between an innovation
community pattern and a learning community pattern. In the Webinar prepa-
ration phase, the core team collaborated as a COIN, whereas in the delivery
phase the speakers and audience collaborated as a CLN.
Figure B.14 illustrates the progress of the communication activities over
time related to preparing and conducting the Webinar. The center of the gure
is a time series plot of the evolution of the GBC and density of organizers and
audience duringthe entire period. The picture at the topleft shows the structure
of the team preparing the Webinar, which is operating as a COIN, with high
1
5
8
Figure B.14. Webinar visualization, with a plot of GBC in the center.
appendix b
159
density and low GBC (arrow 1). The picture at the top right is a screen shot
of the communication pattern during the rst delivery of the Webinar. The
practice leader (dark central dot) is sending and receiving information in a
star structure; the dent in the graph in the center, as pointed out by arrow 2,
identies this phase of highGBC. Duringandafter this rst runof the Webinar,
it is the practice leader who elds most questions. The picture at the lower left
displays the team preparing a rerun of the Webinar, again working as a COIN
and communicating with low GBC (arrow 3). Finally, the screen shot in the
lower right displays the practice coordinator (dark central dot) rerunning the
Webinar, communicating with his audience and answering their questions in
a star structure with high GBC centrality (arrow 4).
Innovation Pattern: Rise and Demise of Startup Cybermap Systems
In sidebar 5.2, Cybermap Systems was introduced as an example of a COIN
that failed. Here, with TeCFlow, we can use the Cybermap example to
illustrate changes in knowledge ow during the three main phases in the life
cycle of a COIN. In this case, TeCFlow was employed to examine the e-mail
archive of the startups founder to visualize and analyze the knowledge ow
and communication patterns among team members. The founders e-mail
archive served as a substitute for the organizational memory of the COIN.
Figure B.15 shows the changes in GBC and density during the critical
periods in the life of the startup. The GBC plot identies the three main
phases during Cybermap Systems existence:
1. Connect: The core group was connected by the founder.
2. Collaborate: The core group worked together as a team.
3. Dissolve: The communication owstalled as the founder was distracted
and withdrew from communication with the rest of the team.
Figure B.16 consists of three snapshots of the communication owmovie
automatically generated by TeCFlow. The snapshots again illustrate the
critical phases in the startup companys life cycle. In picture 1, we see the
initial connect phase, where the founder (the dot in the center) is connecting
the lead programmer (dot close to the founder) and other future core team
members. The founder is contacting his various friends, trying to convince
160
Figure B.15. GBC and density of Cybermap Systems communication flow.
Figure B.16. Three stages in the life of Cybermap Systems.
appendix b
161
them to join his startup. Because these friends are not yet connected among
themselves, the founder acts as a connector and a hub of trust. Picture 2 is the
fully operational startup, with core team members fully connected with each
other and exchanging multiple messages each day. The founder maintains a
central role as the hub of trust, but each core team member communicates
directly with all other core team members. Picture 3 illustrates the dissolve
phase, in this case the beginning of the startups demise. Communication has
stalled; the founder is distracted from running the startup and has withdrawn
from daily communication. He is now located at the teams periphery. While
he is still fully connected, the rest of the core team is exchanging more
messages without including the founder.
In this case, TeCFlow was used retrospectively to shed light on the inner
workings of a startup company. The temporal plot of GBC and the movie of
the communication ow convey deep insights into the rise and demise of this
COIN that would otherwise have been difcult to obtain.
Project Management Pattern: Communication
in a Software Development Team
Our next example analyzes a Swiss e-banking project (discussed in chapter
5), one of the projects delivered by the Deloitte e.Xpert practice. The core
team on Deloitte Consultings side included some 20 consultants, led by a
project manager and a project partner; the client team comprised the client
project manager, a senior manager, andsixsubject matter experts. The analysis
described here focuses on the phase of the project just before go live and the
handover of software to the client. This period featured intense negotiations
between client project management and senior managers of the consulting
practice. There was also a change in Deloittes project management.
Figures B.17 through B.20 are snapshots that illustrate the knowledge
ow within the team during critical phases of this project period. Figure B.17
shows the communication ow as an addition to the original contract was
negotiated, while the technical core team was hard at work implementing
the technical system. On the gures left side, Deloittes legal team forms
a dense cluster, intensively discussing contractual details. Leaders of both
the consulting team and the client project team are in the center of the
structure, communicating with everyone. The technical core team members
appendix b
162
Figure B.17. Contract negotiations with the client.
Figure B.18. Testing of critical programming code.
are forming another COIN at the top of the gure, chiey communicating
among themselves.
During the subsequent testing phase shown in gure B.18, the test coor-
dinator has a centralized role coordinating the developers. The legal team is
still tightly clustered, while the clients (lighter dots) are (too) peripheral. In
appendix b
163
gure B.19, legacy data are converted from the old to the new system. The
database administrator is clearly recognizable in the center of the technical
core team. The legal teamis no longer active, and client and consulting leaders
collaborate somewhat more intensively. Finally, in the handover phase (gure
B.20), client and consulting leaders collaborate closely, while the technical
Figure B.19. Conversion of legacy data.
Figure B.20. Handover of entire application to client.
appendix b
164
Figure B.21. GBC of project team over 150 days.
core teamforms a separate cluster. Collaboration between the client technical
team and the consulting technical team is low.
Figure B.21 illustrates the main activities during these project phases.
The project has generally highly democratic and low GBC. Major events
become peaks in the curve. Project management takeover by the new leader
can be identied clearly, as his initially more centralized communication style
leads to a spike in GBC. The next peak in the curve comes froma knowledge-
gathering workshop actively coordinated jointly by the project leader and the
knowledge manager. Organizing handover as well as the nal pullout of the
consultants led to the next peaks.
Figure B.22 illustrates the communication patterns of the different actors
in the project. We can see that the consulting project leaders receive more
messages than they send, and that they are also the most active senders. The
programmers receive more messages than they send, while the administrative
staff members send more messages that they receive. This is fairly typical
behavior for a commercial software project.
The ndings from the TeCFlow analysis provide a fast and convenient
way to pinpoint weaknesses in the project management process and can
assist project managers to improve efciency and communication ow. In this
particular case, we can see several areas where TeCFlow could have helped.
appendix b
165
Figure B.22. Contribution index of different people in the e-banking project.
The old sales manager still held key client relationships even after
the introduction of the new sales manager. The new sales manager
could have been much more active in leveraging the connections of
the retiring manager in developing his relationships.
The most active customers were not known and, therefore, could
not adequately be taken care of by Deloitte.
Summary
The ve examples in this appendix showthat TeCFlowanalysis of the owof
knowledge offers a fast, convenient, and valuable way to discover different
170
Figure B.26. GBC of account management team.
Figure B.27. Contribution index of account management team.
appendix b
171
phases inthe life cycle of anonline community. TeCFlowconveys insights that
would be much more expensive to obtain by other means, and the TeCFlow-
based approach makes it possible to nd periods of lowand high democracy
(i.e., GBC) and to identify periods of high productivity and information
dissemination. While this tool needs to be complemented by other contextual
cues (such as interviews with community members and a content analysis of
the messages exchanged) to obtain a full understanding of the activities, it is
clearly a software tool with powerful applications for uncovering COINs and
putting them to benecial use in their hosting organizations.
Appendix C introduces knowledge ow optimization as a method for
converting organizations into COINs. Knowledge ow optimization utilizes
TeCFlow to nd emerging clusters of people working together as innovation
teams. It assists in creating a culture for COINs to convert team members
into creators, collaborators, and communicators, thereby greatly improving
organizational creativity, quality, and efciency.
This page intentionally left blank
APPENDIX C
Knowledge Flow Optimization (KFO)
One summer morning in New Hampshire, I was hiking up Mount Lafayette
from the Appalachian Mountain Clubs Greenleaf Hut. The weather forecast
had predicted nice, sunny weather. Suddenly, though, the sky became covered
with dark clouds, and a few drops of rain began to fall. Because I had trusted
the weather forecast, I was wearing only light clothes and had no heavy
raincoat. Within ve minutes, I was soaking wet.
As I got wetter and wetter, I began to reect on the reliability of the
weather forecast. Todays weather forecasting allows for predictions about
weather phenomena based on observations of factors such as atmospheric
pressure, wind speed and direction, precipitation, cloud cover, temperature,
and humidity. In a weather forecasting system, weather patterns consisting of
time series of recently collected data are fed into a computer model to predict
the weather patterns of the next 5 to 15 days.
The parallels between weather forecasting and knowledge ow optimiza-
tion (KFO) are striking. In KFO, communication patterns consisting of time
series of collected communication data provide insights into complex group
dynamics and make it possible to predict future group behavior (gure C.1).
As with weather patterns used to predict sunshine and thunderstorms, com-
munication ows allow for predicting positive and negative developments
in groups of people. This requires a delicate combination of mathematical
modeling, experience, and intuition. KFO is extremely valuable as an early
173
appendix c
174
Figure C.1. Weather patterns and communication patterns allow for predicting
the future.
warningsystem, showinghigh-pressure systems, impendingstorms, andother
relationships in groups that are difcult to anticipate through other means. It
offers insights into organizational dynamics. By analyzing and aligning busi-
ness processes and knowledge ow, organizations get a unique opportunity
to increase the productivity of knowledge workers through greater creativity,
efciency, and quality.
Business process reengineering forever changed the way companies do
business, introducing a process focus and streamlining structured business
processes. KFO does the same for unstructured, knowledge-intensive inno-
vation processes. By visualizing the owof knowledge, making it transparent,
and reengineering its ow, organizations and individuals become more cre-
ative, innovative, andresponsive tochange. KFOoffers companies a chance to
complement their business process maps and organizational charts with much
more uid maps of relationships (gure C.2). By making the communication
ow transparent, KFO can make existing business processes more efcient,
allowing organizations to make better use of people by unburying them from
conventional multilayer hierarchies. By establishing exible ad hoc work
1
7
5
Figure C.2. KFO aligns business processes and organizational structure through communication flow analysis.
appendix c
176
ows based on communication and relationships, people can become more
efcient in their roles, which in turn will lead to greater individual motivation.
Knowledge ow links people across organizational and process bound-
aries, and analysis of this knowledge ow gives an early indication of hidden
innovation. Nascent creative communities can be uncovered by visualizing
intra- and interorganizational communication patterns; business processes
and organizational structures can then be aligned with the actual commu-
nication ow. KFO introduces a framework for the temporal analysis and
visualization of communication patterns using the temporal communication
ow analysis (TeCFlow) software tool introduced in appendix B. TeCFlow
makes it possible for organizations to uncover traces of emerging collab-
orative innovation networks (COINs) by visualizing their communication
ows.
There are four steps to KFO (gure C.3). First, communities are analyzed
by TeCFlow and by interviewing organization members. Second, existing
communication patterns are assessed on the degree to which they resem-
ble the collaborative knowledge network (CKN) framework (i.e., innova-
tion diffusion through the double helix communication pattern) and the three
dimensionsinteract, connect, shareof online behavior discussed in chap-
ter 6 and appendix A. Third, the organization is optimized by applying the
principles of COINs explained in earlier chapters. Fourth, communication
ow is monitored continuously using TeCFlow.
KFOStep 1: Visualize Virtual Communities to Find COINs
The rst step in KFO is to visualize the communication ow and make
it transparent. The TeCFlow visualizer introduced in appendix B offers a
convenient way to uncover communication ows using e-mail and phone
logs, Web access archives, and chat session transcripts.
1
As gure C.4 shows,
the TeCFlow tool addresses both the macro and the micro levels of teams.
On the group or macro level, interactive movies and the plot of changes of
group betweenness centrality (GBC) allow observers easily and quickly to
distinguish different communication patterns over the lifetime of an online
community. On the individual or micro level, interactive movies and the
contribution index make it possible to extract individual communication
patterns.
177
Figure C.3. Knowledge flow optimization.
178
Figure C.4. Macro- and micro-level communication pattern analysis. (Top panel)
Macro level---communication patterns of groups . (Bottom panel) Micro
level---communication patterns of individuals.
appendix c
179
By generating interactive communication movies, we can identify typical
communication patterns of COINs, collaborative learning networks (CLNs),
and collaborative interest networks (CINs). Each of these types of communi-
ties has a distinctive structure; COINs have the strongest small-world, scale-
free structure. The three TeCFlow snapshots in gure C.5 illustrate how a
Figure C.5. Snapshots of TeCFlow communication flow movie to identify COINs/
CLNs/CINs.
appendix c
180
COIN, a CLN, and a CIN grow over time and become identiable (here, we
use how the CKN concept was developed and spread out in the Deloitte
e.Xpert practice as our example). The top picture shows the COIN core team
becoming clearly recognizable as a cluster in the center of the community.
The middle picture shows the COIN growing into a CLN as more members
join the periphery. The bottom picture shows the last TeCFlow snapshot
of the same movie, illustrating the CIN structure of the entire community.
The Deloitte e.Xpert CIN includes the COIN and CLN, but also has many
other members in the e.Xpert practice who have not yet been infected by the
CKN virus.
The three-step TeCFlow analysis described in appendix B can also be
used to analyze the roles of individuals in virtual communities; to help iden-
tify the experts within the organization; and to identify who has the most
decision-making inuence and who the innovators, collaborators, and com-
municators are. Calculating and plotting communication frequency and con-
tribution index with TeCFlow is a quick and convenient way to nd central
COIN members.
In addition to mining communication archives, input for TeCFlow can be
obtained through periodic online questionnaires administered to members of
virtual communities or even to outsiders.
2
Figure C.6 is a sample online form
on which community members can indicate the frequency of their e-mail,
phone, and face-to-face communication.
3
KFOStep 2: Benchmark COINs and Core Contributors
Once a COIN and its core contributors have been identied, performance
should be measured to nd areas for improving knowledge worker produc-
tivitywhich can be a real challenge. There are three primary ways to assess
a COINs performance:
1. Evaluate the COINs output in nancial terms. Once a COIN has
developed a tangible product, the value of that product can be
measured. For the Deloitte e.Xpert virtual consulting practice, we
could measure its cost and revenue against operating a conventional
consulting practice. For DaimlerChryslers e
3
e-procurement project
(see chapter 5), realized savings could be compared to project costs.
appendix c
181
Figure C.6. Web survey form.
The drawback of this method is that it can be difcult to correlate
nancial results with the efforts of a COIN.
4
2. Ask external experts to rate the COINs performance. Collecting the
subjective evaluations of a large number of outsiders is another way
to assess the impact of COINs. This can be done through online
questionnaires, with the obvious drawback that such a method can
absorb a lot of time and resources.
3. Assess objective measures. Using the CKN framework of innovation
diffusion and the three dimensions of online behavior discussed in
chapter 6 and appendix A, it is possible to measure the speed of
innovation diffusion, how long it takes for the COINs innovation
to be accepted, and the level of that acceptance.
5
The basic idea is
to correlate objective measures of COIN success such as the value
appendix c
182
generated by the COIN with social network structure and individual
communication behavior.
High-performing virtual teams exhibit distinctive networking properties.
Social networking analysis metrics can be computed both for groups and
individuals.
6
On the group level, these metrics identify typical patterns of
COINs, CLNs, and CINs. Productive COINs use the CKN ecosystem to col-
laborate and to disseminate information. By exploiting correlations between
the structure of the social network and metrics such as GBC and density,
potentially high-performing group structures such as COINs can be found.
An effective COIN will exhibit a combination of high density with low GBC
over sustained periods of time.
By measuring the position of individuals in networks, we can draw
conclusions about their performance. Using the contribution index function
of TeCFlow, the three dimensions of online behavioragain, interactivity,
connectivity, and sharingcan be measured for each team member; this will
determine the individuals role as creator, collaborator, communicator, or
knowledge expert.
KFOStep 3: Redesign and Optimize
COIN creation is a uid processa COIN cannot be mandated into action.
The progression of COINgrowth is similar to chemical crystallization, where
a nurturing liquid and a crystallization germ are enough to start the process.
Just as a crystal grows from the germ on its own by adding more molecules
through chemical attraction forces, the trigger for COIN members to join the
community stems from their own motivation. Growing the largest crystals
with the most desirable properties requires a thorough understanding of the
procedure and ingredients of the crystallization process. The same principle
applies to the COIN growth process (although in the case of COINs, it may
be better to have multiple, smaller COINs).
Seven success factors are involved in bringing a COIN to fruition. It is
not necessary to accomplish them in sequence, but each one is critical.
1. Establish swarm creativity and give up central control.
2. Nurture the critical roles of creator, communicator, and collaborator.
appendix c
183
3. Establish distributed trust.
4. Establish a common code of ethics.
5. Establish a small-world structure of high connectivity, interactivity,
and knowledge-sharing.
6. Set up a collaborative Web workplace.
7. Know when to change the organizational structure.
Senior executives in organizations that host COINs must create an envi-
ronment in which their COINs have the opportunity to achieve all seven of
these critical success factors, and in which all COIN members can become
the best at playing whichever COIN role they ll. The potential benet is
enormous: imagine getting an intrinsically motivated group of people to deal
with a tough problem not for promotion or a pay rise, but out of fascination
Sidebar C.1
Making COINs Happen
Throughout this book, our focus has been on the existence of COINs (even if they
are hidden from view) and on the conviction business organizations would be
wise to foster a culture supportive of COINs in order to reap their benefits. Most
cases presented in chapter 5 describe COINs that seem to have emerged in the
course of work---that is, that seem to have happened naturally.
This section departs from that focus to take a bold leap and propose that
there are actual steps that can be taken to make COINs happen.
The ideas behind how to create COINs actively are not yet fully worked out.
Companies are trying out the COIN concepts and KFO, and their experiences
are forming the basis for further analysis. In my own project work, I am testing
whether it is possible to generate swarm creativity (not simply create the best
conditions for swarm creativity) by taking great pains to ensure that only
intrinsically self-motivated people who share the same DNA (with respect
to being potential COIN members) comprise a team. This also involves actively
ensuring that each location has a hub of trust, and making sure, as well, that the
hubs of trust get to know each other and establish mutual trust in a face-to-face
meeting (or meetings).
Each of the seven critical success factors in KFO step 3 involves action, and
therein lies the bold leap. Can COINs actively be created? The evidence thus far
strongly suggests yes.
appendix c
184
with the problem and a shared vision for coming up with an innovative
solution.
7
Critical Success Factor 1: Establish Swarm
Creativity and Give Up Central Control
Only if senior executives are willing to give up centralized control and
ready to trust the self-organizing capabilities of COINs will they harvest
the full benets of COINs. Recruiting the brightest and best, and then letting
them loose on their own, requires high levels of condence and managerial
constraint. The same goes for investors, who expect centralized control over
processes to ensure a return. But if investors want to leverage innovation
by COINs, they will have to take a leap of faith and embrace swarmcreativity
and self-organization.
8
Critical Success Factor 2: Nurture Creators,
Collaborators, and Communicators
Growing a CKN ecosystem requires skills different than those managers
typically learn in business schools. Managers need to emphasize putting
together well-functioning teams with members who have the right genes to
swarm around their leaders.
Ateamof creators, collaborators, and communicators does not need much
centralized control. Rather, it needs strategic guidance and an environment
that offers the cultural and technological prerequisites (mutual trust and nec-
essary Internet collaboration tools at the disposal of every team member) for
a CKN ecosystem. Once core team members have been identied, the team
should be managed as an X-team (see chapter 5), with members changing
their roles according to project needs. Accepting a continuously changing
team, in which knowledge experts become creators, collaborators, and com-
municators, and in which core team members withdraw to become external
experts, allows a exible online community to grow, continually building
high mutual trust and a distinct online identity.
Managers who insist on substituting centralized control for trust will drive
COINs underground. There, they may continue to grow, but the chances
appendix c
185
are greatly reduced that the hosting organization will be able to reap the
benets.
Critical Success Factor 3: Establish Distributed Trust
For COINs to succeed, it is essential to create and maintain trust among COIN
members. This is achieved most efciently by meeting face-to-face, but can
happen remotelyalthough it will take more time. If a group of people works
together over long distances and an extended period, and if team members
consistently deliver high-quality work, trust will develop remotely.
Sharing project-relevant knowledge and maintaining a transparent work
environment are important elements of building and preserving a high level of
trust, as is developing a shared vision and common goals. When teammembers
participate in shaping the vision and goals of the community, the result is sub-
stantial buy-in
9
(which builds trust). The process of adjusting and ne-tuning
goals should also be conducted in public, with an opportunity for all COIN
members to collaborate so that the entire community identies with the goals.
Critical Success Factor 4: Establish a Common Code of Ethics
When people are fully committed and intrinsically motivated, they will
implicitly abide by a common ethical code when they work together. They
will apply the Tao of COINs (see chapter 4) without having to learn it. Leaders
who demonstrate respect for the individual and who obey the Golden Rule
will serve as role models for the entire community.
Nevertheless, making the Tao of COINs explicit, and customizing it to the
local environment, will clarify the rules for new COIN members and acceler-
ate the process by which they learn the right behavior from COIN elders.
Critical Success Factor 5: Establish a Small-World Structure
of High Connectivity, Interactivity, and Knowledge Sharing
ACKNecosysteminteroperates as a scale-free, small-world network in which
teams are mutually connected by leaders acting as gateways and hubs of trust.
Having strong leaders act as local hubs of trust will build an efcient, exible,
appendix c
186
and robust network of global scalability. Local leaders should know and trust
each other. Under ideal circumstances, leaders at different locations locally
recruit new core team members whom they already know and trust.
The fastest way to build trust is to meet other core team members face-
to-face. If this is too expensive, a collaborative Web workspace provides the
technical means for COIN members to work together over the Internet and
to develop condence in the skills of other COIN members through reliable
cooperation over time. At present, the e-mail component of the collaborative
Web workspace is the primary means of communication for a geographically
distributed COIN.
Critical Success Factor 6: Set Up a Collaborative Web Workspace
By combining simple Web-based collaboration tools such as e-mail, blogs,
wikis, and Web conferencing, an organization can build a collaborative
Web workspace to support COINs. COINs should establish a persistent
organizational memory on the Web that is easily accessible to all members
to share documents.
10
Weblogs, wikis, and mailing lists offer a cheap and
efcient way to establish that group memory.
11
A protocol for successful e-mail communication in COINs has a few
simple rules: leaders of the community must be easily accessible by e-mail,
and COIN members should respond quickly to messages (or indicate their
temporary unavailability if they are unable to respond in a timely manner).
Videoconferencing, which approximates the intimacy of face-to-face
meetings at lower cost, has become increasingly common at large compa-
nies. Smaller companies and individuals can also rent videoconferencing
capabilities. But while useful for a meeting now and then if people cannot
be brought together in one location, frequent use of videoconferencing is
no substitute for building trust through face-to-face meetings, and hence is
not worth the price. Web conferencing, though, is a cost-efcient substitute
for bringing people together who have already worked together in the past
and want to meet regularly. By using document-sharing software such as
Placeware, meeting participants can share one screen virtually, while public
and private chat windows allow side communications. Low-cost Webcams,
or even a conference call, can be used to discuss what is shown on the shared
screen. For addressing a large group, Webcasts can be usefulfor instance, if
appendix c
187
Sidebar C.2
An Ideal Wiki Application
Wikipedia is the best-known wiki application, allowing people with very few
computer skills to collaborate and share knowledge on the Web. The figure
presented here illustrates how easy it is to change an entry in Wikipedia. The top
panel shows the Wikipedia entry for Tim Berners-Lee as seen by the reader. By
clicking on the edit this page tab, a user can open the entry for editing (the
middle panel). Clicking on the discussion tab (bottom panel) brings up the
page on which authors can sort out disagreements about the text on the main
entry page.
Wikipedia entry for Tim Berners-Lee.
appendix c
188
a CEO wants to address all of a companys employees, or to transmit keynote
speeches of a conference.
Critical Success Factor 7: Know When to Change
to a Conventional Organizational Structure
The nal key to success is to recognize when it is time for a change. Once
a COIN has developed a commercial product, the organizational form of
the COIN should be changed into a project team or a business unit. At
the same time, the CKN ecosystem developing around the COIN should
be leveraged according to the double helix of innovation. Disseminating
innovations from COIN to CLN to CIN is a cost-efcient way to gain a large
audience for the work products of the COIN. The learning network of the CLN
is instrumental in recruiting new core team members. The interest network
is the perfect mechanism for disseminating the innovation. Lurkers at the
periphery of the CIN are the agents of knowledge dissemination, spreading
the seeds of acceptance for the new product. At the same time, the CIN is the
incubator of new innovation, creating an environment in which new COINs
can crystallize and grow.
KFOStep4: Monitor COINs
Maintaining a culture for COINs is not a one-time activity. Once the knowl-
edge ow of the organization has been optimized according to the seven
critical success factors above, the ingredients for a COINs-supportive culture
should be constantly supervised, gauged, adjusted, and remixed if necessary.
Hence, step 4 of KFO is the continuous monitoring and nurturing of a culture
for COINs.
As a communications cockpit, TeCFlowprovides an excellent mechanism
for monitoring communication owbased on e-mail, chat, phone, and Web ac-
cess trafc. In this way, an organization can produce weather forecasts that
reveal high-pressure systems and impending storms. The communication
patterns of such forecasts identify emerging COINs, pinpointing clusters of
high communication trafc as indicators of new innovation, as well as of
groups that need closer attention.
appendix c
189
Realignment of business processes and knowledge ow offers a great
chance to individuals to become better creators, collaborators and communi-
cators. But there are also steps that individuals can take themselves to become
more effective COIN members.
Howto Be a Successful COINMember
Linux is an ecosystem, and in this ecosystem there is fast
growth vegetation and slow growth vegetation. The fast
growth vegetation are the people who took what had already
been done by Unix, and without changing its design they
copied it while making coding improvements.
Then there are those who look at Linux, and see something
just barely begun that needs a complete overhaul. These are
the slow growth vegetation.
12
External fame and reputation, or recommendation by a trusted community
leader, can jump-start the initial position and career of a new COIN member.
A trusted position within a COIN, though, must be earned over time. Senior-
level COIN membership comes only to those who deliver high-quality work
over an extended period. Joining a COINbecause of an intrinsic motivation is
the road to sustainable success, and tackling hard problemsslow-growth
vegetationis the best way to gain an online reputation in the community.
As Hans Reiser puts it, exemplary COIN members are people who chose
trying to create a better society as their lifes work at a substantial cost in
personal income.
13
A good online reputation acquired through hard work is easy to lose.
By acting sloppy, being unreliable, or breaking the ethical code, a COIN
member will quickly diminish his or her reputation. The way to become a
successful COIN member and sustain that success is to follow the rules of
swarm creativity, ethical behavior, and distributed trust.
KFOApplications
Knowledge-intensive organizations that optimize the ow of knowledge re-
alize strategic advantages. Consulting rms, software development projects,
appendix c
190
project management of large projects, mergers and acquisitions, and sales
forces are among the application areas where KFOis ideally suited to improve
organizational creativity, quality, and effectiveness.
Get the most strategic value frommergers and acquisitions. The Daimler-
Chrysler e
3
initiative (see chapter 5) is a prime example of howanorganization
can derive tremendous strategic and nancial value by redesigning its knowl-
edge ow. While the initiative began as a project to reengineer the business
processes of the merged Daimler and Chrysler procurement departments, it
quickly became much more. The initiative turned the entire business model
of procurement upside down and created a multitier online marketplace for
suppliers of auto parts, unleashing tremendous value for the enterprise. The e
3
initiative teamat DaimlerChrysler operatedas a genuine COIN, withmembers
fromboth sides of the Atlantic adhering to the Tao of COINs, creating the new
solution in true swarmcreativity, and collaborating in a highly efcient small-
world networking structure with the senior project members acting as hubs
of trust. The DaimlerChrysler case is an exemplary illustration of the strate-
gic advantages organizations in merger situations can expect by the internal
transparency, consistency, and meritocracy of COINs.
Optimize research and development. Research and development organi-
zations that redesign their knowledge ow so that it operates as a network
of COINs embedded in a CKN ecosystem can expect substantial advantages.
The development of the CKN service offering for Deloitte Consulting (see
chapter 5) illustrates how an innovative new product was developed by a
COIN that was recruiting new members from its surrounding CLN while
using its CIN as a sounding board and sales and marketing mechanism.
Streamline project management. The e-banking project analysis in ap-
pendix Bshows howapplying TeCFlowto redesigning the owof knowledge
would have improved the efciency of a multimillion-dollar project by at least
20 percent. Monitoring project management communication for better qual-
ity of project output translates into substantial savings. The KFO approach
offers myriad communication-related benets: it greatly reduces communi-
cation failures among project members; it converts one-way communication
into two-way dialogues; and it reveals core contributors as well as lurkers.
Changing project culture to a COIN-based approach makes the teamwork
more efcient, unlocking the creative potential of neglected team members.
Visualizing the ow of knowledge makes it easier to nd good ideas within
the organization.
appendix c
191
Improve the sales process. The sales force communication pattern anal-
ysis at the end of appendix B shows how applying KFO improves the ef-
ciency and productivity of sales and marking. TeCFlow makes it possible
to nd who is most productive and unproductive on a sales and marketing
staff.
14
This page intentionally left blank
NOTES
Introduction
1. Gladwell, The TippingPoint: HowLittle Things CanMake aBigDifference (2002).
2. Now Sir Tim Berners-Lee, he was knighted in 2003 by Queen Elizabeth II of the
United Kingdom for his contributions to the creation of the World Wide Web.
3. Bush, As We May Think (1945).
Chapter 1
1. Hamel, Waking Up IBM: How a Gang of Unlikely Rebels Transformed Big Blue
(2000).
Chapter 2
1. Adapted from Bonabeau, Dorigo, and Theraulaz, Swarm Intelligence: From Nat-
ural to Articial Systems (1999).
2. See Wikipedia at www.wikipedia.org, Encarta at encarta.msn.com, and Ency-
clopaedia Britannica at www.britannica.com.
3. Alex Halavais, a University of Buffalo communication professor, conducted
an experiment to address concerns about the accuracy and scientic validity of an
encyclopedia that relies entirelyonvolunteers for its updates. He enteredmistakes into13
Wikipedia entries to see if his hacker attack would be detected and to gauge how long
it would take the volunteers to x the mistakes. His optimistic expectation was that the
errors would be xed within 24 hours; however, he was surprised to nd that it took the
Wikipedia cyberteamonly a fewhours to correct all his mistakes. For more information,
see alex.halavais.net/news/index.php?p=794. Halavaiss ndings have been conrmed
193
notes
194
by IBMresearchers who looked at controversial articles, such as the entry on abortion.
They found that while entries were sometimes purposefully altered or even deleted, it
usually just took only minutes for the malevolent alterations to be xed. For more
information, see www.research.ibm.com/history/index.htm.
4. See Challet and Du, Closed Source versus Open Source in a Model of Software
Bug Dynamics (2003); Raymond, The Cathedral and the Bazaar (1999); Stark, The
Organizational Model for Open Source (2003); Paulson, Succi, and Eberlein, An
Empirical Study of Open-Source and Closed-Source Software Products (2004).
5. Christensen, The Innovators Dilemma: The Revolutionary National Bestseller
That Changed the Way We Do Business (2000).
6. An argument can be made that the Web and Linux were nondisruptive, as there
was already a system called gopher available which was very similar to the Web.
UNIX was also available well before Linux. But by making the Web and Linux freely
available, Tim Berners-Lee and Linus Torvalds were able to rally widespread support
for their causes, which in turn had a disruptive effect on their respective industries.
7. Von Hippel, The Sources of Innovation (1997 [1988]).
8. Chesbrough, Open Innovation: The New Imperative for Creating and Proting
from Technology (2003).
9. Shneiderman, Leonardos Laptop (2002).
10. Licklider, Man-Computer Symbiosis (1960).
11. Digital Systems Research Center, The Tao of IETF: A Guide for New Attendees
of the Internet Engineering Task Force (1994).
Chapter 3
1. In his seminal work, Francis Fukuyama describes the basic lack of trust outside
of family ties in Confucian Chinese society. See Fukuyama, Trust: The Social Virtues
and the Creation of Prosperity (1995).
2. Contrast Marco Polos writings with the writings of Sir John Mandeville. Making
himself the focus of his book, Mandeville claimed to have been a mercenary for the
Sultan of Egypt and to have served the Emperor of China in his war against a Southern
Asian king. Descriptions of the Cyclops and people with the heads of dogs are some
of the less outrageous of Mandevilles fantastic claims. See Mandeville, Travels of Sir
John Mandeville (1984 [1356]).
3. Hancock, Thom-Santelli, and Ritchie, Deception and Design: The Impact of
Communication Technologies on Lying Behavior (2004).
4. Joinson, Understanding the Psychology of Internet Behaviour: Virtual Worlds,
Real Lives (2003).
5. Gates, Business @ Speed of Thought, Using a Digital Nervous System (1999).
6. It should be noted that leaders of Microsofts competition, such as Sun Microsys-
tems CEO Scott McNealy, are very critical of this approach to innovation and accuse
notes
195
Microsoft of outright intellectual property robbery. When Microsoft created the same
look and feel as the Macintosh user interface through 189 identical elements in its
Windows operating system, Apple took the company to courtand lost.
7. Saxenian, Regional Advantage: Culture and Competition in Silicon Valley and
Route 128 (2000). Apple has learned this lesson, while Microsoft is increasingly closing
its own interfaces to other software vendors. For example, Microsoft is currently
involved in a lawsuit in the European Union, having been sued by vendors of operating
system add-ons such as digital audio and video players. Competitors, including Real-
Audio, claimMicrosoft is gaining an unfair advantage by tightly bundling its own Media
Player software with its operating system, repeating a marketing gimmick that worked
perfectly in the past to stop PC buyers from using competitor Netscapes Web browser.
This shows Microsoft is becoming extremely selective in its approach to transparency
and information sharing.
8. SHAREs success thus far is impressive. In its rst four years, SHARE linked
hundreds of entrepreneurs, startups, and researchers. In 2003, it organized more than
60 events and hosted more than 4,500 visitors. SHARE also makes heavy use of the
Web to support collaboration among geographically dispersed innovative knowledge
workersfor example, by offering live Web transmission of its networking events.
Chapter 4
1. Kanter, Commitment and Community: Communes and Utopias in Sociological
Perspective (1972), p. 73.
2. Digital Systems Research Center, The Tao of IETF: A Guide for New Attendees of
the Internet Engineering Task Force (1994).
3. Gillmor, In the Wild West of the Internet, There Are Good Guys and Bad Guys
(2003).
4. Rawls, A Theory of Justice (1999), p. 86.
5. Watts, Tao: The Watercourse Way (1977).
6. Digital Systems Research Center, The Tao of IETF.
7. Moon and Sproull, The Essence of Distributed Work: The Case of the Linux
Kernel (2000).
8. Putnam, Bowling Alone: The Collapse and Revival of American Community
(2000).
9. See http://smallworld.columbia.edu/.
Chapter 5
1. In 2004, Deloitte Consulting merged with the consulting services of Deloitte &
Touche, and the entire rm has been renamed Deloitte.
notes
196
2. The breadth of this effort was very wide, encompassing consultants at ofces in
Amsterdam, Brussels, Copenhagen, Frankfurt, Helsinki, Lisbon, London, Luxembourg,
Madrid, Milan, Munich, Oslo, Paris, Rome, Stockholm, Vienna, and Zurich.
3. Williams and Cockburn, Agile Software Development: Its about Feedback and
Change (2003).
Chapter 6
1. For example, Web conferencing tools such as Placeware (purchased by Microsoft
in 2003 and renamed LiveMeeting) allow the participants to see who the other listen-
ers are.
2. Research has demonstrated that the quality of group performance of a distributed
teamis directly proportional to the degree of knowledge-sharing among group members.
See Cummings and Cross, Structural Properties of Work Groups and Their Conse-
quences for Performance (2003).
Appendixes
1. Licklider and Taylor, The Computer as a Communication Device (1968).
Appendix A
1. These communities are sometimes called practice communities or knowledge
stewarding communities. See Wenger, McDermott, and Snyder, Cultivating Commu-
nities of Practice (2002).
2. Gardner and Gardner, Your Sneak Peek into Motley Fool Stock Advisor (2002).
3. CKN is the name of a Deloitte Consulting service offering based on an earlier
and much less fully developed version of the collaborative knowledge network concept
described in this appendix. At the time Deloitte offered this service, for instance, there
was no distinction between COINs, CINs, CLNs, and CKNs. How Deloittes CKN
service was created is actually a great example of how a COIN works, as described later
in this section.
4. See www.deloitte.com/dtt/cda/doc/content/collaborativeknowledge.pdf.
5. Deloittes CKNinitiative won considerable recognition fromindustry analysts and
in the press. Articles about this work appeared in German, Austrian, Finnish, English,
and United States newspapers. More important, the tools, concepts, and methodologies
developed by the CKN initiative opened doors to potential clients across Europe, the
United States, and Southeast Asia. As I was leaving Deloitte at the end of 2002, CKN-
related projects were under way in Finland, Switzerland, Austria, the United States, the
United Kingdom, Singapore, and elsewhere.
notes
197
Appendix B
1. TeCFlow, a visual tool for temporal communication ow analysis, is currently
being developed as part of a CKNjoint project between the MITCenter for Coordination
Science and the Center for Digital Strategies at Dartmouths Amos Tuck School of
Business Administration.
2. One could reasonably argue that measuring e-mail is insufcient for approximating
social ties, and that this should be complemented by phone logs and transcripts of face-
to-face interaction. In experiments by others, though, e-mail has been conrmed to be
a generally reliable indicator and proportional to the other interactions.
3. In this particular working group, the W3C board appointed some leaders. As we
shall see, when some of the appointed leaders failed to fulll their leadership roles, other
non-appointed people stepped in to lead the groupin typical COIN-like fashion.
Appendix C
1. To nd the hidden COINs in organizations, a combination of mining communica-
tion archives with TeCFlow and interviewing executives and members of the organiza-
tion is recommended.
2. The main advantage of online questionnaires is they make it relatively easy to
collect and combine communicationassuming users respond (which is the potential
drawback). By contrast, data such as e-mail, phone logs, and chat transcripts can be
collected automatically without requiring that users enter data on their own.
3. Respecting individual privacy is of the utmost importance for the success of a
COIN communication assessment. To obtain the buy-in of the study population, there
must be assurance that the identity of individuals will be protected as data are collected,
and that analysis results will be reported only in an anonymous format.
4. Admittedly, it can be very difcult to measure nancially a COINs output
but it is crucial. Some past efforts to do so have looked at productivity increases by
measuringrevenue generatedincomparisontocosts before andafter KFO. This approach
is very complicated, though, and so some simpler process-specic measures are advised:
savings in reduced work hours, newrevenue generated, reduced inventory costs, reduced
production costs, etc.
5. Consider, for example, that fax machines took 100 years to be accepted and used
by a sizable portion of the population; e-mail took 10 years; and the Web took ve years.
6. Bulkley and Van Alstyne, Why Information Should Inuence Productivity
(2004). See also Cummings and Cross, Structural Properties of Work Groups and Their
Consequences for Performance (2003).
7. The innovationsoften newproductsthat COINmembers create must meet two
related criteria: (a) they must be virtually transportable from one work location to the
other, and (b) to be virtually transportable, they must demand little local context.
notes
198
8. COIN innovation has ourished for the past 20 years in the creation of new
software, but today products as diverse as cars, airplanes, drugs, and movies are
developed virtually on computer monitors. Unlike software, though, these products still
require massive investments in production (e.g., factories, laboratories, studios). Hence
the need for investors to embrace the COINs concept, too.
9. It is also important to secure the buy-in of external sponsors and the organizations
top management as soon as possible. COIN members participate because they are
intrinsically motivated, but they also depend on organizational acceptance. The best
way to secure organizational acceptance is to recruit a fully committed sponsor high up
in the organization. This is not a trivial task: the goals of a COIN are often, at rst, of
only peripheral interest to the hosting organization. The key to success is to persevere,
perhaps even enlisting the support of external promoters who have the attention of top
management of the hosting organization.
10. For example, the group-relevant part of a COINs internal e-mail communication
can easily be stored as a mailing list.
11. Weblogs, which have come to be known as blogs, are Web-based online diaries
where anyone who can type can share thoughts. Blog readers can post comments on
what they read. Wikis take this concept even further, allowing everyone to edit the
contents of a Web page. Tools such as twiki (a user-friendly version of a wiki available
for free at www.twiki.org/) allowfor sharing all types of documents and for coordinating
collaboration and workow on documents through check-in and check-out. While word
processing tools such as Microsoft Word allow users to compare and merge different
versions of text documents, these simpler Web-based tools are entirely sufcient for
coauthoring large documents.
12. Reiser, interview (2003).
13. Ibid.
14. Research has demonstrated that high-performing sales force members communi-
cate more with external people than do average or low performers; that high performers
make greater use of communication technologies for their work; and, surprisingly, that
there is no correlation between performance and overall volume of communication. This
means that very active communicators are not necessarily high performers. Addition-
ally, there is a positive correlation between individual performance and short e-mail
messages. By applying insights such as these to the communication patterns extracted
by TeCFlow, the individual communication behavior of sales and marketing teams can
be optimized for superior performance. See Bulkley and van Alstyne, Does E-Mail
Make White Collar Workers More Productive? (2004).
REFERENCES
Ancona, D., Bresman, H., and Kaeufer, K. (2002). The Comparative Advantage of
X-Teams. MIT Sloan Management Review 43:3, 3339.
Bonabeau, E., Dorigo, M., and Theraulaz, G. (1999). SwarmIntelligence: FromNatural
to Articial Systems. Santa Fe Institute Studies in the Sciences of Complexity. New
York: Oxford University Press.
Bulkley, N., and van Alstyne, M. (2004). Does E-Mail Make White Collar Workers
More Productive? In Proceedings of the North American Association for Compu-
tational Social and Organization Science (NAACSOS), June 2729, Pittsburgh, PA.
Available online at www.casos.cs.cmu.edu/events/conferences/2004 proceedings/
Bulkley Nathaniel.pdf.
Bulkley, N., andvan Alstyne, M. W. (2004). WhyInformation Should Inuence Produc-
tivity. In The Network Society: A Cross-Cultural Perspective, ed. M. Castells. Chel-
tenham, U.K.: Edward Elgar. Available online at http://ssrn.com/abstract=518242.
Bush, V. (1945). As We May Think. The Atlantic Monthly 176:1, 101108.
Challet, D., and Du, Y. L. (2003). Closed Source versus Open Source in a Model of
Software Bug Dynamics. Preprint. Available online at
http://arxiv.org/abs/cond-mat/0306511.
Chesbrough, H. W. (2003). Open Innovation: The New Imperative for Creating and
Proting from Technology. Boston: Harvard Business School Press.
Christensen, C. (2000). The Innovators Dilemma: The Revolutionary National Best-
seller That Changed the Way We Do Business. New York: HarperCollins.
Cormen, T., Leiserson, C., andRivest, R. (1990). IntroductiontoAlgorithms. Cambridge,
Mass.: The MIT Press.
Cummings, J., and Cross, R. (2003). Structural Properties of Work Groups and Their
Consequences for Performance. Social Networks 25:3, 197210.
199
references
200
Digital Systems Research Center. (November 1994). The Tao of IETF: A Guide for New
Attendees of the Internet Engineering Task Force. RFC 1718. Updated August 2001.
Available online at www.ietf.org/tao.html.
Fukuyama, F. (1995). Trust: The Social Virtues and the Creation of Prosperity. New
York: Free Press.
Gardner, D., and Gardner, T. (July 22, 2002). Your Sneak Peek into Motley Fool Stock
Advisor. Motley Fool Stock Advisor [online only]. Available online at
http://investorplace.com/free/mfsa free 002A.php.
Gates, W. (1999). Business @ Speed of Thought, Using a Digital Nervous System. New
York: Warner Books.
Gillmor, D. (September 28, 2003). In the Wild West of the Internet, There Are Good
Guys and Bad Guys. San Jose Mercury News. Available online at
www.siliconvalley.com/mld/siliconvalley/business/columnists/6881523.htm.
Gladwell, M. (2002). The Tipping Point: How Little Things Can Make a Big Difference.
Boston: Back Bay Books.
Gloor, P., Dynes, S., and Lee, I. (1993). Animated Algorithms. CD-ROM. Cambridge,
Mass.: The MIT Press.
Goleman, D. (1997). Emotional Intelligence: Why It Can Matter More Than IQ. New
York: Bantam Books.
Hamel, G. (July/August 2000). Waking Up IBM: How a Gang of Unlikely Rebels
Transformed Big Blue. Harvard Business Review, 137146.
Hancock, J. T., Thom-Santelli, J., and Ritchie, T. (2004). Deception and Design: The
Impact of Communication Technologies on Lying Behavior. In Proceedings of the
SIGCHI Conference on Human Factors in Computing Systems, 129134. New York:
ACM Press.
Joinson, A. N. (2003). Understanding the Psychology of Internet Behaviour: Virtual
Worlds, Real Lives. Hampshire, U.K.: Palgrave Macmillan.
Kanter, R. M. (1972). Commitment and Community: Communes and Utopias in Socio-
logical Perspective. Cambridge, Mass.: Harvard University Press.
Leary, W. E. (April 14, 2004). Better Communication Is NASAs Next Frontier. New
York Times, A24.
Licklider, J. C. R. (March 1960). Man-Computer Symbiosis. IRE Transactions on
Human Factors in Electronics HFE-1, 411.
Licklider, J. C. R., and Taylor, R. (April 1968). The Computer as a Communication
Device. Science and Technology, pages unavailable.
Malone, T. W. (2004). The Future of Work: How the New Order of Business Will Shape
Your Organization, Your Management Style, and Your Life. Boston: Harvard Business
School Press.
Mandeville, J. (1984 [1356]). Travels of Sir John Mandeville. London: Penguin Books.
Moon, J. Y., and Sproull, L. (November 2000). The Essence of Distributed Work: The
references
201
Case of the Linux Kernel. First Monday 5:11. Available online at
www.rstmonday.dk/issues/issue5 11/moon/.
Paasivaara, M., Lassenius, C., and Pyysi ainen, P. (2003). Communication Patterns and
Practices in Software Development Networks. Helsinki, Finland: Helsinki University
of Technology.
Paulson, J., Succi, G., and Eberlein, A. (April 2004). An Empirical Study of Open-
Source and Closed-Source Software Products. IEEE Transactions on Software
Engineering 30:4.
Putnam, R. (2000). Bowling Alone: The Collapse and Revival of American Community.
New York: Simon & Schuster.
Rawls, J. (1999 [1971]). A Theory of Justice. Cambridge, Mass.: Harvard University
Press.
Raymond, E. (1999). The Cathedral and the Bazaar. Sebastopol, Calif.: OReilly Media.
Available online at www.catb.org/esr/writings/cathedral-bazaar/.
Reiser, H. (June 18, 2003). Interview on Slashdot.org. Available online at
http://interviews.slashdot.org/article.pl?sid=03/06/18/1516239.
Saxenian, A. (2000). Regional Advantage: Culture and Competition in Silicon Valley
and Route 128. Cambridge, Mass.: Harvard University Press.
Shneiderman, B. (2002). Leonardos Laptop. Cambridge, Mass.: The MIT Press.
Stark, M. (July 7, 2003). The Organizational Model for Open Source. HBS Working
Knowledge [online database]. Available online at
http://workingknowledge.hbs.edu/pubitem.jhtml?id=3582&t=technology.
Von Hippel, E. (1997 [1988]). The Sources of Innovation, rev. ed. New York: Oxford
University Press.
Wasserman, S., and Faust, K. (1994). Social Network Analysis: Methods and Applica-
tions. Cambridge: Cambridge University Press.
Watts, A. (1977). Tao: The Watercourse Way. New York: Random House.
Wenger, E., McDermott, R., and Snyder, W. (2002). Cultivating Communities of Prac-
tice. Boston: Harvard Business School Press.
Williams, L., and Cockburn, A. (June 2003). Agile Software Development: Its about
Feedback and Change. IEEE Computer 36:6, 3943.
This page intentionally left blank
ADDITIONAL RESOURCES
Allen, T. (1977). Managing the Flow of Technology. Cambridge, Mass.: MIT Press.
Barabasi, A. (2002). Linked: The NewScience of Networks. Cambridge, Mass.: Perseus.
Batagelj, V., and Mrvar, A. (1998). PajekProgram for Large Network Analysis.
Connections 21:2, 4757. Available online at
http://vlado.fmf.uni-lj.si/pub/networks/doc/pajek.pdf.
Bonabeau, E., and Meyer, C. (May 2001). Swarm Intelligence: A Whole New Way to
Think about Business. Harvard Business Review 5, 107114.
Borgatti, S., Everett, M., and Freeman, L. C. (1992). UCINET IV, Version 1.0. Columbia,
SC: Analytic Technologies.
Borgatti, S. P., and Everett, M. G. (1999). Models of Core/Periphery Structures. Social
Networks 21: 375395.
Burt, R. S. (1992). Structural Holes. Cambridge, Mass.: Harvard University Press.
Cailliau, R., and the World Wide Web Consortium (W3C). (1995). A Little History
of the World Wide Web from 1945 to 1995. HTML timeline. Available online at
http://www.w3.0rg/History.html.
Chandler, A. D. (1977). The Visible Hand. Cambridge, Mass.: Harvard University Press.
Chesbrough, H. W. (July 2003). A Better Way to Innovate. Harvard Business Review
81:7, 1213.
Cohen, D., and Prusak, L. (2001). In Good Company. How Social Capital Makes
Organizations Work. Boston: Harvard Business School Press.
Cross, R., Nohria, N., and Parker, A. (Spring 2002). Six Myths about Informal
Networksand How to Overcome Them. MIT Sloan Management Review 43:3,
6775.
Cross, R., and Prusak, L. (June 2002). The People Who Make Organizations Goor
Stop. Harvard Business Review 80:6, 105112.
203
additional resources
204
Cusumano, M. A. (1998). Microsoft Secrets: How the Worlds Most Powerful Software
Company Creates Technology, Shapes Markets, and Manages People. New York:
Simon & Schuster.
Davenport, T., and Prusak, L. (2000). Working Knowledge. Boston: Harvard Business
School Press.
Deloitte Consulting. (2001). Collaborative Knowledge Networks: Driving Workforce
Performance through Web-Enabled Communities. Available online at
http://www.dc.com/obx/pages.php?Name=AllResearch sub ckn.
Dodds, P. E., Muhamad, R., and Watts, D. J. (August 2003). An Experimental Study
of Search in Global Social Networks. Science 301:8, 827829.
Ebel, H., Mielsch, L., and Bornholdt, S. (February 12, 2002). Scale-Free Topology of
E-Mail Networks. Physical Review E 66, 035103R.
Ellickson, R. C. (1991). Order without Law. Cambridge, Mass.: Harvard University
Press.
Farhoomand, A. F., Ng, P. S. P., and Conley, W. L. (April 2003). Building a Successful
E-Business: The FedEx Story. Communications of the ACM 46:4, 8489.
Freeman, L. C., and Freeman, S. C. (1980). A Semi-Visible College: Structural Ef-
fects of Seven Months of EIES Participation by a Social Networks Community.
In Electronic Communication: Technology and Impacts, M. J. MacNaughton and
M. M. Henderson, eds., 7785. AAAS Symposium 52. Washington, D.C.: American
Association for the Advancement of Science.
Gasser, L., and Scacchi, W. (October 15, 2003). Continuous Design of Free/Open Source
Software. Workshop Report and Research Agenda. Irvine, Calif., and Urbana, Ill.:
UCI-UIUC Workshop on Continuous Design of Open Source Software.
Girvan, M., and Newman, M. E. J. (December 7, 2001). Community Structure in Social
and Biological Networks. Proceedings of the National Academy of Sciences USA
99, 78217826.
Gloor, P. (December 1991). CYBERMAPYet Another Way of Navigating in Hyper-
space. In Proceedings of the Third Annual ACMConference on Hypertext, 107121.
New York: ACM Press.
Gloor, P. (1992). AACEAlgorithm Animation for Computer Science Education. In
Proceedings of the 1992 IEEEWorkshop on Visual Languages, 2531. Seattle, Wash.:
IEEE Computer Society.
Gloor, P. (2000). Making the E-Business Transformation: Sharing Knowledge in the
E-Business Company. London: Springer.
Gloor, P. (November 2002). Collaborative Knowledge Networks. eJETA: The Elec-
tronic Journal for e-Commerce Tools and Applications 1:2. Available online at
www.ejeta.org.
Gloor, P., and Dynes, S. (1998). CybermapVisually Navigating the Web. Journal
of Visual Languages and Computing 9, 319336.
additional resources
205
Gloor, P., Laubacher, R., Dynes, S., and Zhao, Y. (2003). Visualization of Interaction
Patterns in Collaborative Knowledge Networks for Medical Applications. In Pro-
ceedings of the Tenth International Conference on Computer-Human Interaction, vol.
3: Human-Centered Computing: Cognitive, Social, and Ergonomic Aspects, 981985.
Gloor, P., Laubacher, R., Dynes, S., and Zhao, Y. (November 2003). Visualization of
Communication Patterns in Collaborative Innovation Networks: Analysis of Some
W3C Working Groups. In Proceedings of the Twelfth International Conference on
Information and Knowledge Management, 5660. New York: ACM Press.
Gloor, P., and Uhlmann, P. (1999). The Impact of E-Commerce on Developing Coun-
tries. Presentation at the Third Annual Conference on Business Information Systems
(BIS), Poznn, Poland.
Granovetter, M. (May 1978). The Strength of Weak Ties. American Journal of
Sociology 78:6, 13601380.
Greenspun, P. (1999). Using the ArsDigita Community System. ArsDigita Systems
Journal [online only]. Available online at
http://eveander.com/arsdigita/asj/using-the-acs.
Greve, A. (2004). Creativity in Social Networks: Combining Knowledge in Innova-
tions. In Creativity and Problem-Solving in the Context of Business Management.
A Festschrift in Honor of Geir Kaufmann for His 60-Year Anniversary, 143163.
Bergen, Norway: Fagbokforlaget.
Guimera, R., Danon, L., Diaz-Guilera, A., Giralt, F., andArenas, A. (2003). Self-Similar
Community Structure in Organizations. Physical Review E 68, 065103R.
Hagel, J., and Armstrong, A. (1997). Net Gain: Expanding Markets through Virtual
Communities. Boston: Harvard Business School Press.
Jarvenpaa, S. L., Knoll, K., and Leidner, D. E. (1998). Is Anybody Out There?
Antecedents of Trust in Global Virtual Teams. Journal of Management Information
Systems 14:4, 2964.
Kleiner, A. (2002). Karen Stephensons Quantum Theory of Trust. strategy+business
29, 02406R.
Kratzer, J., Leenders, R. T. A. J., and van Engelen, J. M. L. (2004). Stimulating the
Potential: Creativity and Performance in Innovation Teams. Journal of Creativity
and Innovation Management 13, 6370.
Larner, J. (1999). Marco Polo and the Discovery of the World. New Haven, Conn.: Yale
University Press.
Lee, G. K., and Cole, R. E. (October 25, 2000). The Linux Kernel Development as a
Model of Knowledge Development. Working Paper. Berkeley, Calif.: Haas School of
Business, University of California, Berkeley.
Leenders, R. T. A. J., van Engelen, J. M. L., and Kratzer, J. (2003). Virtuality,
Communication, and New Product Team Creativity: A Social Network Perspective.
Journal of Engineering and Technology Management 20, 6992.
additional resources
206
Lueg, C., and Fisher, D. 2003. From Usenet to CoWebs, Interacting with Social Infor-
mation Spaces. London: Springer.
Malone, T. W., and Crowston, K. (1994). The Interdisciplinary Study of Coordination.
ACM Computing Surveys 26:1, 87119.
Malone, T. W., Crowston, K. G., and Herman, G., eds. (2003). Organizing Business
Knowledge: The MIT Process Handbook. Cambridge, Mass: MIT Press.
Malone, T. W., and Laubacher, R. (SeptemberOctober 1998). The Dawn of the
E-Lance Economy. Harvard Business Review 76:5, 144152.
Malone, T. W., Laubacher, R. J., and Scott Morton, M. S., eds. (2003). Inventing the
Organizations of the 21st Century. Cambridge, Mass.: MIT Press.
Meyerson, D., Weick, K. E., and Kramer, R. M. (1996). Swift Trust and Temporary
Groups. In Trust in Organizations: Frontiers of Theory and Research, R. M. Kramer
and T. R. Tyler, eds., 166195. Thousand Oaks, Calif.: Sage.
Ogger, G. (1978). Kauf dir inen Kaiserdie Geschichte der Fugger. Munich, Germany:
DroemerKnaur.
OMahony, S. (2003). Guarding the Commons: How Community Managed Software
Projects Protect Their Work. Research Policy 32, 11791198.
OMahony, S. (in press). Developing Community Software in a Commodity World.
Chapter 8 in Frontiers of Capital: Ethnographic Reections on the New Economy,
M. Fisher and G. Downey, eds. Social Science Research Council.
Orlikowski, W., and Yates, J. (1994). Genre Repertoire: The Structuring of Commu-
nicative Practices in Organizations. Administrative Science Quarterly 39, 541574.
Padgett, J., and Ansell, C. (May 1993). Robust Action and the Rise of the Medici,
14001434. American Journal of Sociology 98, 12591319.
Pyysiinen, P., Paasivaara, M., and Lassenius, C. (2003). Coping with Social Complexity
in Distributed Software Development Projects. Helsinki, Finland: Helsinki University
of Technology.
Schein, E. H., withKampas, P. J., DeLisi, P., andSonduck, M. (2003). DECIs Dead, Long
Live DEC: The Lasting Legacy of Digital Equipment Corporation. San Francisco,
Calif.: Berrett-Koehler.
Seligman, A. B. (1998). Trust and Sociability: On the Limits of Condence and Role
Expectations. American Journal of Economics and Sociology 57:4, 391404.
Shepherd, G., and Shepherd, G. (June 25, 2002). The Family in Transition: The
Moral Career of a New Religious Movement. Presentation at the 2002 CESNUR
Conference, Salt Lake City, Utah.
Siau, K., and Shen, Z. (April 2003). Building Customer Trust in Mobile Commerce.
Communications of the ACM 46:4, 9194.
Tuckman, B. W. (1965). Developmental Sequence in Small Groups. Psychological
Bulletin 63, 38499.
Tuomi, I. (2003). Networks of Innovation. New York: Oxford University Press.
additional resources
207
Tyler, J., Wilkinson, D., and Huberman, B. A. (2003). Email as Spectroscopy: Au-
tomated Discovery of Community Structure within Organizations. In Communities
and Technologies, 8196. Deventer, the Netherlands: Kluwer. Available online from
HP Laboratories at http://www.hpl.hp.com/shl/papers/email/index.html.
Van der Smagt, T. (2000). Enhancing Virtual Teams: Social Relations vs. Communi-
cation Technology. Industrial Management and Data Systems 100:4, 148156.
Varghese, G., and Allen, T. (Spring 1993). Relational Data in Organizational Settings:
An Introductory Note for Using AGNI and Netgraphs to Analyze Nodes, Relation-
ships, Partitions and Boundaries. Connections 16:12.
Von Hippel, E. (June 2002). Horizontal Innovation Networksby and for Users. Work-
ing Paper 436602. Cambridge, Mass.: MIT Sloan School of Management. Available
online at http://web.mit.edu/evhippel/www/Publications.htm.
Von Krogh, G., Ichijo, K., and Nonaka, I. (2000). Enabling Knowledge Creation: How
to Unlock the Mystery of Tacit Knowledge and Release the Power of Innovation. New
York: Oxford University Press.
Watts, D. (1999). Small Worlds. Princeton, N.J.: Princeton University Press.
Watts, D. J., and Strogatz, S. H. (1998). Collective Dynamics of Small-World Net-
works. Nature 393, 440442. Available online at
http://tam.cornell.edu/SS nature smallworld.pdf.
Weber, M. (1947). The Theory of Economic Organization. New York: The Free Press.
Weber, M. (1978). Economy and Society. Berkeley, Calif.: University of California Press.
Weber, S. (2004). The Success of Open Source. Cambridge, Mass.: Harvard University
Press.
Yates, J. (1989). Control through Communication: The Rise of System in American
Management. Baltimore, Md.: Johns Hopkins University Press.
This page intentionally left blank
INDEX
Advanced Research Projects Agency, 42
agile software development, 109
Alinghi, 10
Altos, 41
Andreessen, Marc, 7
animated algorithms, 38
apache.org, 129
ARPANET, 43
AT&T Bell Labs, 32
Athey, Robin, 137
Barabasi, Albert-Laszlo, 85
Bell Labs, 32
Benko, Cathy, 134
Berners-Lee, Tim, 5, 6, 131
Bertarelli, Ernesto, 10
betweenness centrality, 150
Bircher, Fritz, 34
blogs, 198
Bonabeau, Eric, 20
Burchett, Jack, 34
Bush, Vannevar, 5
business process reengineering, 174
Cailliau, Robert, 6, 131
CERN, 6
CINs. See Collaborative Interest Networks
CKNs. See Collaborative Knowledge
Networks
Clinton, Bill, 44
CLN. See Collaborative Learning Network
Cocoon, 27
code of ethics,71, 86, 185
COINs
benchmarking, 180
benets, 12
collaboration and communication, 92
communication technology, 115
core group, 146
creation, 92
critical success factors, 182188
denition, 4
discovering, 142
DNA of, 49
ending, 139
genetic code, 53
patterns of members, 122
redesign, 182
technology, 93, 115
transformation, 140
Collaborative Innovation Networks. See
COINs
Collaborative Interest Networks, 127140
discovering, 142
Collaborative Knowledge Networks,
127140
Brown Bag, 138
conception, 137139
double helix,131
e.Xperts as, 134
ripple effect, 130
support software, 132
209
index
210
Collaborative Learning Network, 127140
discovering, 142
collaborative Web workspace, 119121,
186
collaborator, 122, 147, 184
commitment-building process, 72, 73
Common Object Request Broker
Architecture, 111
commune, 72
communication cockpit, 188
communication ow analysis, 175
communication patterns
analysis in COINs, 145
of innovation, 154, 159
learning, 156
of project management, 161
of sales force, 166
communication technology, 115
communicator, 122, 147, 184
Community of Practice, 196
Confucian China, 194
connector, 44
consistency, 79, 82
contribution index, 146149
CORBA. See Common Object Request
Broker Architecture
core/periphery structure, 150
Covisint, 94
creator, 122, 147, 184
Cybermap, 107
innovation pattern of, 159
da Vinci, Leonardo, 36
DaimlerChrysler
e-extended Enterprise (e
3
), 9, 30, 93,
190
Dartmouth Center for Digital Strategies at
Tuck, 197
Dean, Howard, 19
Debian, 54
Deloitte Consulting, 99
business sector, 100
Collaborative Knowledge Network
initiative, 137139,196
e-business council, 135
e-Champion, 100, 102
e.Xpert, 99, 166, 190
service line, 100
Deloitte Research, 137
density, 150
Deshpande Center, 68
double helix
of Collaborative Knowledge Networks,
131
e-banking, 104
e-Champion, 100, 102
Eclipse, 27
CKN, as a, 28
e-extended Enterprise (e
3
), 9,30, 93, 190
ethical code of, 95
small world structure of, 95
e-mail, 116
Encyclopedia Britannica, 24
Engelbart, Douglas, 5, 40, 116
ethical code, 71, 86, 95, 185
e.Xpert, 99, 166, 190
hub of trust, 101
Collaborative Knowledge Networks as,
134
face-to-face interaction, 116
Fairchild Semiconductor, 41
fax, 116
ame war, 60
Fugger, Jakob, 50
Fukuyama, Francis, 86
Gamma, Erich, 27
Gates, Bill, 41, 63
Gladwell, Malcolm, 3, 44
Grannell, Douglas, 23
group betweenness centrality, 149
150
gunpowder, 51
Gutenberg, Johannes, 52
Hoenigsberg, David, 23
Homo sapiens, 49
honesty, 59
hub of trust, 88, 101
IBM
Eclipse, 27
Research Labs, 32
innovation
diffusion, 130
disruptive, 32
grassroots, 33
sustaining, 31
Intel Research, 57
Internet, creation, 43
index
211
Internet Engineering Task Force, 45, 74
Interoperability Service Interface, 27, 29,
110
Jobs, Steve, 41
Kanter, Rosabeth Moss, 72
Kay, Alan, 41
Kenyan Association for the Advancement
of Computer Technology, 61
knowledge expert, 147
knowledge ow, 16
knowledge ow optimization, 173191
four steps of, 176189
Krause, Reinhold, 35
Lampson, Butler, 41
learning network, 49
Leonardo da Vinci, 36
Licklider, J.C.R. 42, 125
Linux, 25, 130, 189, 194
Luther, Martin, 52
Malone, Thomas W., 3, 137
Mandeville, Sir John, 194
many-to-many multicast, 117
Marco Polo, 56
Marmier, Pascal, 68
maven, 44
Mazzochi, Stefano, 27
Medici, Lorenzo, 51
Melzi, Francesco, 36
Memex, 5
meritocracy, 67, 79, 82
Microsoft, 41
digital nervous system, 63
Internet Explorer, 64
Microsoft Encarta, 24
Milgram, Stanley, 84
Milovich, Dimitrije, 34
MIT Center for Coordination Science,
197
MobileJet, 34
Morse, Samuel, 116
Mosaic, 7
Motley Fool, 128, 129
multicast, 117
NASA, 80
Neanderthal, 49
Nelson, Ted, 5
Netscape, 7
Navigator, 64
Network Working Group, 45
nongovernmental organizations, 80
Noyce, Robert, 41
Ojanga, Tom, 61
oneFish, 129
online behavior, 118119, 121
online questionnaire, 181, 197
oNLine System (NLS), 6
online trust, 86
open source
software, 2529
motivation, 29
Pacioli, Luca, 36
PlanetLab, 57
Polo, Marco, 56
Popes opera, 23
Poppen, Sherman, 34
Poverty Liberation Project, 62
Putnam, Robert, 82
Rawls, John, 75
Reagan, Ronald , 44
Red Hat, 131
Reiser, Hans, 189
ripple effect, of Collaborative Knowledge
Networks, 130
roles, discovering in COINs, 145
Rothschild, Mayer Amschel, 52
Route 128, 66
Sachs, Paul, 44
Salai, Andrea, 36
Saxenian, Annalee, 66
scale free
hubs of trust, 88
network, 85
SHARE, 67, 195
Silicon Valley, 66
Sketchpad, 41
slashdot.org, 129
small world, 71, 185
COIN structure of a, 142
hubs of trust in, 88
network, 84
Smith, Gareth Dylan, 23
snowboarding, 34
snurfer, 34
index
212
sociability, spontaneous, 86
social capital, 80, 81
World Bank view, 83
SOCNET, 78
software patent, 65
structural hole, 15
Sun Microsystems, 194
SuSE, 131
Sutherland, Ivan, 40
swarm creativity, 184
denition, 22
swarm intelligence, 20
swarming, 19
Swiss House for Advanced Research and
Education, 67, 195
Swiss private bank, 104
Tao
of COINs, 75, 77, 86
of Internet Engineering Task Force, 76
Taoism, 76
Taylor, Robert, 41, 125
TeCFlow, 102, 141171, 190, 197
knowledge ow analysis, for, 149
Temporal Communication Flow Analysis.
See TeCFlow
Tennenhouse, David, 57
Thacker, Charles, 41
theory of justice, 75
tipping point, 3, 44
Torvalds, Linus, 25, 130
Trade Point, 113
transparency, 58, 67, 79, 82
trust, 86
hubs of, 88
UBS, 61, 108
Bank Wide Web, 108
Interoperability Service Interface, 27,
110
Unctad, 113
United Nations Conference on Trade and
Development, 113
VA Linux, 131
VisiCalc, 64
Von Arb, Christoph, 68
W3C (World Wide Web Consortium), 129,
147
Wal-Mart, 31
Watts, Duncan, 85
weather forecast, 173
Webinar, 157
weblog, 198
Wikipedia, 24, 187, 193
Wikiquette, 25
World Bank, 83
World Wide Web, 5, 32, 194
creation of, 46
Xerox Palo Alto Research Center, 32
X-team, 97
denition, 98